38 template <
typename I,
typename K,
typename C>
Valid convolution, with multiple kernels.
Same convolution, with multiple kernels.
conv_type
Enumeration describing the different types of convolution.
Definition: conv_select.hpp:20
Contains selectors for convolution implementations.
constexpr size_t conv1_parallel_threshold_conv
The mimum output size before considering parallel convolution.
Definition: threshold.hpp:71
Root namespace for the ETL library.
Definition: adapter.hpp:15
context & local_context()
Return the configuration context of the current thread.
Definition: context.hpp:50
constexpr bool is_parallel
Indicates if the expressions and implementations can be automaticallly parallelized.
Definition: config.hpp:59
auto parallel(Expr &&expr) -> parallel_expr< detail::build_type< Expr >>
Create a parallel expression wrapping the given expression.
Definition: wrapper_expression_builder.hpp:79
constexpr size_t size(const E &expr) noexcept
Returns the size of the given ETL expression.
Definition: helpers.hpp:108
bool select_parallel(const I &, const K &kernel, C &&conv)
Test if ETL should run in parallel for the conv of I and K in C.
Definition: conv_select.hpp:39
Contains selectors for convolution implementations.
constexpr bool parallel_support
Indicates if support for parallelization is integrated into the framework.
Definition: config.hpp:51
auto serial(Expr &&expr) -> serial_expr< detail::build_type< Expr >>
Create a serial expression wrapping the given expression.
Definition: wrapper_expression_builder.hpp:66
constexpr size_t conv1_parallel_threshold_kernel
The mimum kernel size before considering parallel convolution.
Definition: threshold.hpp:72
Contains selectors for convolution implementations.
Full convolution, with multiple kernels.