How to use Boost: serialization to save Eigen :: Matrix

Hello, I have a code that implements libeigen2 to compute eigenvectors. Now I want to use boost :: serialization to save information for later retrieval. From the tutorial example, I came up with the following code!

class RandomNode { private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & is_leaf_; ar & depth_; ar & num_classes_; ar & num_features_; // Split node members ar & random_feature_indices_; ar & random_feature_weights_; ar & threshold_; ar & leftChild_; ar & rightChild_; } bool is_leaf_; int depth_; int num_classes_; int num_features_; // Split node members VectorXi random_feature_indices_; VectorXd random_feature_weights_; double threshold_; RandomNode* leftChild_; RandomNode* rightChild_; // Methods and so on } 

Now when I try to run this code, I get the following error

 /usr/include/boost/serialization/access.hpp:118:9: error: 'class Eigen::Matrix<double, 10000, 1>' has no member named 'serialize' 

How can I serialize the Eigen :: Matrix class? Is it possible? Thanks in advance.

+4
source share
1 answer

You should read the boost :: serialization documentation on the topic serializable . This basically suggests that types should be primitive or Serializable. The Eigen type is nothing your compiler is trying to tell you. To make Eigen serializable types, you need to implement the following free function

 template<class Archive> inline void serialize( Archive & ar, my_class & t, const unsigned int file_version ) { ... } 

To do this for Eigen, I think you could do something like this template

Here is an example implementation that should work for you:

 #include <fstream> #include <Eigen/Core> #include <boost/archive/text_oarchive.hpp> using namespace Eigen; struct RandomNode { friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & random_feature_indices_; ar & random_feature_weights_; } // Split node members VectorXi random_feature_indices_; VectorXd random_feature_weights_; }; namespace boost { template<class Archive, typename _Scalar, int _Rows, int _Cols, int _Options, int _MaxRows, int _MaxCols> inline void serialize( Archive & ar, Matrix<_Scalar, _Rows, _Cols, _Options, _MaxRows, _MaxCols> & t, const unsigned int file_version ) { size_t rows = t.rows(), cols = t.cols(); ar & rows; ar & cols; if( rows * cols != t.size() ) t.resize( rows, cols ); for(size_t i=0; i<t.size(); i++) ar & t.data()[i]; } } int main() { // create and open a character archive for output std::ofstream ofs("filename"); RandomNode r; r.random_feature_indices_.resize(3,1); // save data to archive { boost::archive::text_oarchive oa(ofs); // write class instance to archive oa << r; // archive and stream closed when destructors are called } } 
+12
source

Source: https://habr.com/ru/post/952299/


All Articles