PLSSVM - Parallel Least Squares Support Vector Machine
2.0.0
A Least Squares Support Vector Machine implementation using different backends.
|
Class specifying a backend independent execution range. More...
#include <execution_range.hpp>
Public Member Functions | |
execution_range (std::initializer_list< std::size_t > grid, std::initializer_list< std::size_t > block) | |
Initialize the grid and block sizes using std::initializer_list s. More... | |
template<std::size_t I, std::size_t J, PLSSVM_REQUIRES((0< I &&I<=3 &&0< J &&J<=3)) > | |
execution_range (const std::array< std::size_t, I > &p_grid, const std::array< std::size_t, J > &p_block) | |
Initialize the grid and block sizes using std::array s. Only available if the number of values specified for the grid and block sizes are greater than zero and less or equal than three. More... | |
Public Attributes | |
std::array< std::size_t, 3 > | grid = { 1, 1, 1 } |
The grid sizes (using the CUDA definition). | |
std::array< std::size_t, 3 > | block = { 1, 1, 1 } |
The block sizes (using the CUDA definition). | |
Class specifying a backend independent execution range.
Holds two members: grid
specifying the grid size and block
specifying the block size using the CUDA definition. Both grid and block must specify at least a one dimensional and at most a three dimensional value used in the kernel invocation.
plssvm::detail::execution_range::execution_range | ( | std::initializer_list< std::size_t > | grid, |
std::initializer_list< std::size_t > | block | ||
) |
Initialize the grid and block sizes using std::initializer_list
s.
If less than three values are specified, fills the missing values with ones. Uses the CUDA definition.
[in] | grid | specifies the grid sizes |
[in] | block | specifies the block sizes |
|
inline |
Initialize the grid and block sizes using std::array
s. Only available if the number of values specified for the grid and block sizes are greater than zero and less or equal than three.
If less than three values are specified, fills the missing values with ones. Uses the CUDA definition.
[in] | p_grid | specifies the grid sizes |
[in] | p_block | specifies the block sizes |