Compute Library
 23.08
DepthfirstDriver< TInput, TWeight, TOutput > Class Template Referenceabstract

#include <depthfirst_driver.hpp>

Collaboration diagram for DepthfirstDriver< TInput, TWeight, TOutput >:
[legend]

Public Member Functions

 DepthfirstDriver (IDepthfirstStrategy *strategy, const DepthwiseArgs &args)
 
size_t get_working_size (unsigned int n_threads) const override final
 
virtual bool supports_direct_padding () const
 

Detailed Description

template<typename TInput, typename TWeight, typename TOutput>
class arm_conv::depthwise::DepthfirstDriver< TInput, TWeight, TOutput >

Definition at line 66 of file depthfirst_driver.hpp.

Constructor & Destructor Documentation

◆ DepthfirstDriver()

DepthfirstDriver ( IDepthfirstStrategy strategy,
const DepthwiseArgs &  args 
)
inline

Definition at line 273 of file depthfirst_driver.hpp.

274  : Parent(args), m_strat(strategy)
275  {
276  }

Member Function Documentation

◆ get_working_size()

size_t get_working_size ( unsigned int  n_threads) const
inlinefinaloverride

Definition at line 278 of file depthfirst_driver.hpp.

279  {
280  return n_threads * this->get_working_size_per_thread();
281  }

◆ supports_direct_padding()

virtual bool supports_direct_padding ( ) const
inlinevirtual

Definition at line 283 of file depthfirst_driver.hpp.

284  {
285  return false;
286  }

The documentation for this class was generated from the following file:
GemmTuner.args
args
Definition: GemmTuner.py:679
strategy
const StratType * strategy
Definition: working_space.hpp:105