Public Member Functions
const bool	fortran_contiguous () const

const Scalar *	data () const

const std::vector< size_t > &	shape () const

template<int Rows, int Cols, int Options>
	operator Eigen::Matrix< Scalar, Rows, Cols, Options > () const

Friends
std::istream &	operator>> (std::istream &is, NumpyInput &data)

Detailed Description

NumpyInput allows the unserialization of a contiguous numpy array from a stream.

Example:

numpy_format::NumpyInput<float> ni;
std::cin >> ni;
MatrixXf x = ni;

Member Function Documentation

template<typename Scalar>

template<int Rows, int Cols, int Options>

ldaplusplus::numpy_format::NumpyInput< Scalar >::operator Eigen::Matrix< Scalar, Rows, Cols, Options > ( ) const

inline

This operator allows the implicit cast of a NumpyInput to a compatible Eigen matrix.

numpy_format::NumpyInput<float> ni;
std::cin >> ni;
MatrixXf x = ni; // this conversion is done with this operator

template<typename Scalar>

friend

Parse the serialized numpy array into a NumpyInput from an input strem.

numpy_format::NumpyInput<float> ni;
std::cin >> ni; // this operator here

Only version 1 of the numpy format is supported but that should not present any problems.

The documentation for this class was generated from the following file: