File regression_matrix_csr.hpp

Functions

std::unique_ptr<ICsrRegressionMatrix> createCsrRegressionMatrix(float32 *values, uint32 *indices, uint32 *indptr, uint32 numRows, uint32 numCols)

Creates and returns a new object of the type ICsrRegressionMatrix.

Parameters:

  • values – A pointer to an array of type float32, shape (numDenseElements), that stores the values of all dense elements explicitly stored in the matrix

  • indices – A pointer to an array of type uint32, shape (numDenseElements), that stores the column indices of all dense elements explicitly stored in the matrix

  • indptr – A pointer to an array of type uint32, shape (numRows + 1), that stores the indices of the first element in indices that corresponds to a certain row. The index at the last position is equal to numDenseElements

  • numRows – The number of rows in the regression matrix

  • numCols – The number of columns in the regression matrix

Returns:

An unique pointer to an object of type ICsrRegressionMatrix that has been created

class ICsrRegressionMatrix : public virtual IRowWiseRegressionMatrix
#include <regression_matrix_csr.hpp>

Defines an interface for all regression matrices that provide row-wise access to the regression scores of individual examples that are stored in a sparse matrix in the compressed sparse row (CSR) format.

Public Functions

inline virtual ~ICsrRegressionMatrix() override