Convert Eigen Matrix to C array

Question

The Eigen library can map existing memory into Eigen matrices.

float array[3];
Map(array, 3).fill(10);
int data[4] = 1, 2, 3, 4;
Matrix2i mat         


        

Answer 1

If the array is two-dimensional, one needs to pay attention to the storage order. By default, Eigen stores matrices in column-major order. However, a row-major order is needed for the direct conversion of an array into an Eigen matrix. If such conversions are performed frequently in the code, it might be helpful to use a corresponding typedef.

using namespace Eigen;
typedef Matrix<int, Dynamic, Dynamic, RowMajor> RowMatrixXi;

With such a definition one can obtain an Eigen matrix from an array in a simple and compact way, while preserving the order of the original array.

From C array to Eigen::Matrix

int nrow = 2, ncol = 3;
int arr[nrow][ncol] =  { {1 ,2, 3},  {4, 5, 6} }; 
Map<RowMatrixXi> eig(&arr[0][0], nrow, ncol);

std::cout << "Eigen matrix:\n" << eig << std::endl;

// Eigen matrix:
// 1 2 3
// 4 5 6

In the opposite direction, the elements of an Eigen matrix can be transferred directly to a C-style array by using Map.

From Eigen::Matrix to C array

int arr2[nrow][ncol];
Map<RowMatrixXi>(&arr2[0][0], nrow, ncol) = eig;

std::cout << "C array:\n";
for (int i = 0; i < nrow; ++i) {
  for (int j = 0; j < ncol; ++j) {
    std::cout << arr2[i][j] << " ";
  }
  std::cout << "\n";
}

// C array:
// 1 2 3 
// 4 5 6 

Note that in this case the original matrix eig does not need to be stored in row-major layout. It is sufficient to specify the row-major order in Map.

Answer 2

You can use the data() member function of the Eigen Matrix class. The layout by default is column-major, not row-major as a multidimensional C array (the layout can be chosen when creating a Matrix object). For sparse matrices the preceding sentence obviously doesn't apply.

Example:

ArrayXf v = ArrayXf::LinSpaced(11, 0.f, 10.f);
// vc is the corresponding C array. Here's how you can use it yourself:
float *vc = v.data();
cout << vc[3] << endl;  // 3.0
// Or you can give it to some C api call that takes a C array:
some_c_api_call(vc, v.size());
// Be careful not to use this pointer after v goes out of scope! If
// you still need the data after this point, you must copy vc. This can
// be done using in the usual C manner, or with Eigen's Map<> class.

Answer 3

You need to use the Map function again. Please see the example here: http://forum.kde.org/viewtopic.php?f=74&t=95457

Answer 4

ComplexEigenSolver < MyMatrix > es;
complex<double> *eseig;
es.compute(H);
es.eigenvalues().transpose();
eseig=(complex<double> *)es.eigenvalues().data();

Answer 5

The solution with Map above segfaults when I try it (see comment above).

Instead, here's a solution that works for me, copying the data into a std::vector from an Eigen::Matrix. I pre-allocate space in the vector to store the result of the Map/copy.

Eigen::MatrixXf m(2, 2);
m(0, 0) = 3;
m(1, 0) = 2.5;
m(0, 1) = -1;
m(1, 1) = 0;

cout << m << "\n";

// Output:
//    3  -1
// 2.5   0

// Segfaults with this code: 
//
// float* p = nullptr;
// Eigen::Map<Eigen::MatrixXf>(p, m.rows(), m.cols()) = m;

// Better code, which also copies into a std::vector:

// Note that I initialize vec with the matrix size to begin with:
std::vector<float> vec(m.size());
Eigen::Map<Eigen::MatrixXf>(vec.data(), m.rows(), m.cols()) = m;

for (const auto& x : vec)
  cout << x << ", ";
cout << "\n";

// Output: 3, 2.5, -1, 0

Answer 6

To convert normal data type to eigen matrix type

  double *X; // non-NULL pointer to some data

You can create an nRows x nCols size double matrix using the Map functionality like this:

  MatrixXd eigenX = Map<MatrixXd>( X, nRows, nCols );

To convert eigen matrix type into normal data type

  MatrixXd resultEigen;   // Eigen matrix with some result (non NULL!)
  double *resultC;        // NULL pointer <-- WRONG INFO from the site. resultC must be preallocated!
  Map<MatrixXd>( resultC, resultEigen.rows(), resultEigen.cols() ) =   resultEigen;

In this way you can get in and out from eigen matrix. Full credits goes to http://dovgalecs.com/blog/eigen-how-to-get-in-and-out-data-from-eigen-matrix/