imbalance_bounded_limit is a strategy_type which decides the number of stored elements per row of the ell part. More...

#include <ginkgo/core/matrix/hybrid.hpp>

Inheritance diagram for gko::matrix::Hybrid< ValueType, IndexType >::imbalance_bounded_limit:

Collaboration diagram for gko::matrix::Hybrid< ValueType, IndexType >::imbalance_bounded_limit:

Public Member Functions
	imbalance_bounded_limit (double percent=0.8, double ratio=0.0001)
	Creates a imbalance_bounded_limit strategy.

size_type	compute_ell_num_stored_elements_per_row (array< size_type > *row_nnz) const override
	Computes the number of stored elements per row of the ell part. More...

auto	get_percentage () const
	Get the percent setting. More...

auto	get_ratio () const
	Get the ratio setting. More...

Public Member Functions inherited from gko::matrix::Hybrid< ValueType, IndexType >::strategy_type
	strategy_type ()
	Creates a strategy_type.

void	compute_hybrid_config (const array< size_type > &row_nnz, size_type ell_num_stored_elements_per_row, size_type coo_nnz)
	Computes the config of the Hybrid matrix (ell_num_stored_elements_per_row and coo_nnz). More...

size_type	get_ell_num_stored_elements_per_row () const noexcept
	Returns the number of stored elements per row of the ell part. More...

size_type	get_coo_nnz () const noexcept
	Returns the number of nonzeros of the coo part. More...

Detailed Description

template<typename ValueType = default_precision, typename IndexType = int32>
class gko::matrix::Hybrid< ValueType, IndexType >::imbalance_bounded_limit

imbalance_bounded_limit is a strategy_type which decides the number of stored elements per row of the ell part.

It uses the imbalance_limit and adds the upper bound of the number of ell's cols by the number of rows.

Member Function Documentation

◆ compute_ell_num_stored_elements_per_row()

template<typename ValueType = default_precision, typename IndexType = int32>

size_type gko::matrix::Hybrid< ValueType, IndexType >::imbalance_bounded_limit::compute_ell_num_stored_elements_per_row ( array< size_type > * row_nnz ) const