Compute Library
 21.02
NEGEMMLowpOffsetContributionKernel Class Reference

Neon kernel used to add the offset contribution after NEGEMMLowpMatrixMultiplyKernel. More...

#include <NEGEMMLowpOffsetContributionKernel.h>

Collaboration diagram for NEGEMMLowpOffsetContributionKernel:
[legend]

Public Member Functions

const char * name () const override
 Name of the kernel. More...
 
 NEGEMMLowpOffsetContributionKernel ()
 Constructor. More...
 
 NEGEMMLowpOffsetContributionKernel (const NEGEMMLowpOffsetContributionKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
NEGEMMLowpOffsetContributionKerneloperator= (const NEGEMMLowpOffsetContributionKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 NEGEMMLowpOffsetContributionKernel (NEGEMMLowpOffsetContributionKernel &&)=default
 Allow instances of this class to be moved. More...
 
NEGEMMLowpOffsetContributionKerneloperator= (NEGEMMLowpOffsetContributionKernel &&)=default
 Allow instances of this class to be moved. More...
 
 ~NEGEMMLowpOffsetContributionKernel ()=default
 Default destructor. More...
 
void configure (ITensor *mm_result, const ITensor *vector_sum_col, const ITensor *vector_sum_row, int32_t k, int32_t a_offset, int32_t b_offset)
 Initialise the kernel's input and output. More...
 
void run (const Window &window, const ThreadInfo &info) override
 Execute the kernel on the passed window. More...
 
- Public Member Functions inherited from ICPPKernel
virtual ~ICPPKernel ()=default
 Default destructor. More...
 
virtual void run_nd (const Window &window, const ThreadInfo &info, const Window &thread_locator)
 legacy compatibility layer for implemantions which do not support thread_locator In these cases we simply narrow the interface down the legacy version More...
 
virtual void run_op (ITensorPack &tensors, const Window &window, const ThreadInfo &info)
 Execute the kernel on the passed window. More...
 
- Public Member Functions inherited from IKernel
 IKernel ()
 Constructor. More...
 
virtual ~IKernel ()=default
 Destructor. More...
 
virtual bool is_parallelisable () const
 Indicates whether or not the kernel is parallelisable. More...
 
virtual BorderSize border_size () const
 The size of the border for that kernel. More...
 
const Windowwindow () const
 The maximum window the kernel can be executed on. More...
 

Static Public Member Functions

static Status validate (const ITensorInfo *mm_result, const ITensorInfo *vector_sum_col, const ITensorInfo *vector_sum_row, int32_t a_offset, int32_t b_offset)
 Static function to check if given info will lead to a valid configuration of NEGEMMLowpOffsetContributionKernel. More...
 

Detailed Description

Neon kernel used to add the offset contribution after NEGEMMLowpMatrixMultiplyKernel.

The computation is performed in-place

This kernel takes a final int32 accumulator value (the output of NEGEMMLowpMatrixMultiplyKernel), and adds to it the offset contribution of matrix A and matrix B in-place.

The final result is:

mm_result[i][k] = mm_result[i][k] + (vector_sum_col[k] * a_offset) + (vector_sum_row[i] * b_offset) + (a_offset * b_offset * k)

Definition at line 46 of file NEGEMMLowpOffsetContributionKernel.h.

Constructor & Destructor Documentation

◆ NEGEMMLowpOffsetContributionKernel() [1/3]

Constructor.

Definition at line 358 of file NEGEMMLowpOffsetContributionKernel.cpp.

Referenced by NEGEMMLowpOffsetContributionKernel::name().

359  : _vector_sum_col(nullptr), _vector_sum_row(nullptr), _mm_result(nullptr), _a_offset(0), _b_offset(0), _k_offset(0), _slide_vector_sum_col(true)
360 {
361 }

◆ NEGEMMLowpOffsetContributionKernel() [2/3]

Prevent instances of this class from being copied (As this class contains pointers)

◆ NEGEMMLowpOffsetContributionKernel() [3/3]

Allow instances of this class to be moved.

◆ ~NEGEMMLowpOffsetContributionKernel()

Default destructor.

Referenced by NEGEMMLowpOffsetContributionKernel::name().

Member Function Documentation

◆ configure()

void configure ( ITensor mm_result,
const ITensor vector_sum_col,
const ITensor vector_sum_row,
int32_t  k,
int32_t  a_offset,
int32_t  b_offset 
)

Initialise the kernel's input and output.

Parameters
[in,out]mm_resultInput tensor containing the result of NEGEMMLowpMatrixMultiplyKernel. Data type supported: S32
[in]vector_sum_colInput row-vector of sums of all the entries in each column of matrix B. Note: vector_sum_col can be a nullptr in case a_offset = 0. Data type supported: same as mm_result
[in]vector_sum_rowInput row-vector of sums of all the entries in each row of matrix A. Note: vector_sum_row can be a nullptr in case b_offset = 0. Data type supported: same as mm_result
[in]kNumber of matrix A columns or Matrix B rows
[in]a_offsetOffset to be added to each element of the matrix A.
[in]b_offsetOffset to be added to each element of the matrix B.

Definition at line 363 of file NEGEMMLowpOffsetContributionKernel.cpp.

References ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, arm_compute::calculate_max_window(), ITensor::info(), Dimensions< T >::num_dimensions(), ITensorInfo::num_dimensions(), Dimensions< T >::set_num_dimensions(), ITensorInfo::set_valid_region(), ITensorInfo::tensor_shape(), and arm_compute::validate_arguments().

Referenced by NEGEMMLowpOffsetContributionKernel::name().

364 {
365  // Perform validate step
366  ARM_COMPUTE_ERROR_ON_NULLPTR(mm_result);
368  vector_sum_col != nullptr ? vector_sum_col->info() : nullptr, // NOLINT
369  vector_sum_row != nullptr ? vector_sum_row->info() : nullptr, // NOLINT
370  a_offset, b_offset)); // NOLINT
371 
372  _vector_sum_col = vector_sum_col;
373  _vector_sum_row = vector_sum_row;
374  _mm_result = mm_result;
375  _a_offset = a_offset;
376  _b_offset = b_offset;
377  _k_offset = a_offset * b_offset * k;
378 
379  // If a_offset == 0, vector_sum_col can be a nullptr
380  if(a_offset != 0)
381  {
382  // Check if vector_sum_col_shape should be slidden or not
383  // Don't slide vector_sum_col_shape along the y dimension if vector_sum_col_shape has just 1 dimension and vector_sum_row_shape more than 1
384  // This scenario can happen when the the matrix multiplication is used to perform a convolution operation
385  _slide_vector_sum_col = vector_sum_col->info()->tensor_shape().num_dimensions() > 1;
386  }
387 
388  // Configure kernel window
389  Window win = calculate_max_window(*mm_result->info(), Steps());
390  Coordinates coord;
391  coord.set_num_dimensions(mm_result->info()->num_dimensions());
392  mm_result->info()->set_valid_region(ValidRegion(coord, mm_result->info()->tensor_shape()));
393  INEKernel::configure(win);
394 }
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
Status validate_arguments(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo *output_stage)
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161

◆ name()

◆ operator=() [1/2]

Prevent instances of this class from being copied (As this class contains pointers)

Referenced by NEGEMMLowpOffsetContributionKernel::name().

◆ operator=() [2/2]

Allow instances of this class to be moved.

◆ run()

void run ( const Window window,
const ThreadInfo info 
)
overridevirtual

Execute the kernel on the passed window.

Warning
If is_parallelisable() returns false then the passed window must be equal to window()
Note
The window has to be a region within the window returned by the window() method
The width of the window has to be a multiple of num_elems_processed_per_iteration().
Parameters
[in]windowRegion on which to execute the kernel. (Must be a region of the window returned by window())
[in]infoInfo about executing thread and CPU.

Reimplemented from ICPPKernel.

Definition at line 404 of file NEGEMMLowpOffsetContributionKernel.cpp.

References ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, ARM_COMPUTE_UNUSED, ITensor::info(), ITensorInfo::num_dimensions(), ITensorInfo::tensor_shape(), IKernel::window(), Dimensions< T >::x(), and Dimensions< T >::y().

Referenced by NEGEMMLowpOffsetContributionKernel::name().

405 {
409 
410  // Check if input is a 3D reinterpretation
411  const bool reinterpret_as_3d = _vector_sum_row != nullptr
412  && _mm_result->info()->num_dimensions() > 1
413  && _mm_result->info()->tensor_shape().y() != _vector_sum_row->info()->tensor_shape().x();
414 
415  run_offset_contribution(window, _mm_result, _vector_sum_col, _vector_sum_row, _a_offset, _b_offset, _k_offset, _slide_vector_sum_col, reinterpret_as_3d);
416 }
virtual size_t num_dimensions() const =0
The number of dimensions of the tensor (rank)
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
T x() const
Alias to access the size of the first dimension.
Definition: Dimensions.h:87
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Definition: Error.h:152
virtual const TensorShape & tensor_shape() const =0
Size for each dimension of the tensor.
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor&#39;s metadata.
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:941
ScaleKernelInfo info(interpolation_policy, default_border_mode, PixelValue(), sampling_policy, false)
T y() const
Alias to access the size of the second dimension.
Definition: Dimensions.h:92
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:205

◆ validate()

Status validate ( const ITensorInfo mm_result,
const ITensorInfo vector_sum_col,
const ITensorInfo vector_sum_row,
int32_t  a_offset,
int32_t  b_offset 
)
static

Static function to check if given info will lead to a valid configuration of NEGEMMLowpOffsetContributionKernel.

Parameters
[in]mm_resultInput tensor containing the result of NEGEMMLowpMatrixMultiplyKernel. Data type supported: S32
[in]vector_sum_colInput row-vector of sums of all the entries in each column of matrix B. Note: vector_sum_col can be a nullptr in case a_offset = 0. Data type supported: same as mm_result
[in]vector_sum_rowInput row-vector of sums of all the entries in each row of matrix A. Note: vector_sum_row can be a nullptr in case b_offset = 0. Data type supported: same as mm_result
[in]a_offsetOffset to be added to each element of the matrix A.
[in]b_offsetOffset to be added to each element of the matrix B.
Returns
a status

Definition at line 396 of file NEGEMMLowpOffsetContributionKernel.cpp.

References ARM_COMPUTE_RETURN_ON_ERROR, and arm_compute::validate_arguments().

Referenced by NEGEMMLowpOffsetContributionKernel::name(), and NEGEMMLowpMatrixMultiplyCore::validate().

398 {
399  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(mm_result, vector_sum_col, vector_sum_row, a_offset, b_offset));
400 
401  return Status{};
402 }
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
Definition: Error.h:204
Status validate_arguments(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo *output_stage)

The documentation for this class was generated from the following files: