Compute Library
 22.11
CLNormalizationLayerKernel Class Reference

Interface for the normalization layer kernel. More...

#include <CLNormalizationLayerKernel.h>

Collaboration diagram for CLNormalizationLayerKernel:
[legend]

Public Member Functions

 CLNormalizationLayerKernel ()
 Constructor. More...
 
 CLNormalizationLayerKernel (const CLNormalizationLayerKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
CLNormalizationLayerKerneloperator= (const CLNormalizationLayerKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 CLNormalizationLayerKernel (CLNormalizationLayerKernel &&)=default
 Default Move Constructor. More...
 
CLNormalizationLayerKerneloperator= (CLNormalizationLayerKernel &&)=default
 Default move assignment operator. More...
 
void configure (const ICLTensor *input, ICLTensor *output, NormalizationLayerInfo norm_info)
 Set the input and output tensors. More...
 
void configure (const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, NormalizationLayerInfo norm_info)
 Set the input and output tensors. More...
 
void run (const Window &window, cl::CommandQueue &queue) override
 Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue. More...
 
BorderSize border_size () const override
 The size of the border for that kernel. More...
 
- Public Member Functions inherited from ICLKernel
 ICLKernel ()
 Constructor. More...
 
cl::Kernel & kernel ()
 Returns a reference to the OpenCL kernel of this object. More...
 
CLKernelType type () const
 Returns the CL kernel type. More...
 
template<typename T >
void add_1D_array_argument (unsigned int &idx, const ICLArray< T > *array, const Strides &strides, unsigned int num_dimensions, const Window &window)
 Add the passed 1D array's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_1D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 1D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_1D_tensor_argument_if (bool cond, unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 1D tensor's parameters to the object's kernel's arguments starting from the index idx if the condition is true. More...
 
void add_2D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 2D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_2D_tensor_argument_if (bool cond, unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 2D tensor's parameters to the object's kernel's arguments starting from the index idx if the condition is true. More...
 
void add_3D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 3D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_4D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 4D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_5D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 5D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_3d_tensor_nhw_argument (unsigned int &idx, const ICLTensor *tensor)
 Add the passed NHW 3D tensor's parameters to the object's kernel's arguments by passing strides, dimensions and the offset to the first valid element in bytes. More...
 
void add_4d_tensor_nhwc_argument (unsigned int &idx, const ICLTensor *tensor)
 Add the passed NHWC 4D tensor's parameters to the object's kernel's arguments by passing strides, dimensions and the offset to the first valid element in bytes. More...
 
virtual void run_op (ITensorPack &tensors, const Window &window, cl::CommandQueue &queue)
 Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue. More...
 
virtual void run_composite_op (ITensorPack &tensors, const Window &window, cl::CommandQueue &queue, const experimental::dynamic_fusion::ClExecutionDescriptor &exec_desc)
 The execution is carried out through run_op method. But the run_op method needs to be extended to include ClExecutionDescriptor as now LWS GWS tuning will be separated from the IKernel. More...
 
template<typename T >
void add_argument (unsigned int &idx, T value)
 Add the passed parameters to the object's kernel's arguments starting from the index idx. More...
 
void set_lws_hint (const cl::NDRange &lws_hint)
 Set the Local-Workgroup-Size hint. More...
 
cl::NDRange lws_hint () const
 Return the Local-Workgroup-Size hint. More...
 
void set_wbsm_hint (const cl_int &wbsm_hint)
 Set the workgroup batch size modifier hint. More...
 
cl_int wbsm_hint () const
 Return the workgroup batch size modifier hint. More...
 
const std::string & config_id () const
 Get the configuration ID. More...
 
void set_target (GPUTarget target)
 Set the targeted GPU architecture. More...
 
void set_target (cl::Device &device)
 Set the targeted GPU architecture according to the CL device. More...
 
GPUTarget get_target () const
 Get the targeted GPU architecture. More...
 
size_t get_max_workgroup_size ()
 Get the maximum workgroup size for the device the CLKernelLibrary uses. More...
 
template<unsigned int dimension_size>
void add_tensor_argument (unsigned &idx, const ICLTensor *tensor, const Window &window)
 
template<typename T , unsigned int dimension_size>
void add_array_argument (unsigned &idx, const ICLArray< T > *array, const Strides &strides, unsigned int num_dimensions, const Window &window)
 Add the passed array's parameters to the object's kernel's arguments starting from the index idx. 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...
 
const Windowwindow () const
 The maximum window the kernel can be executed on. More...
 
bool is_window_configured () const
 Function to check if the embedded window of this kernel has been configured. More...
 

Static Public Member Functions

static Status validate (const ITensorInfo *input, const ITensorInfo *output, NormalizationLayerInfo norm_info)
 Static function to check if given info will lead to a valid configuration of CLNormalizationLayerKernel. More...
 
- Static Public Member Functions inherited from ICLKernel
static constexpr unsigned int num_arguments_per_3d_tensor_nhw ()
 Returns the number of arguments enqueued per NHW 3D Tensor object. More...
 
static constexpr unsigned int num_arguments_per_4d_tensor_nhwc ()
 Returns the number of arguments enqueued per NHWC 4D Tensor object. More...
 
static constexpr unsigned int num_arguments_per_1D_array ()
 Returns the number of arguments enqueued per 1D array object. More...
 
static constexpr unsigned int num_arguments_per_1D_tensor ()
 Returns the number of arguments enqueued per 1D tensor object. More...
 
static constexpr unsigned int num_arguments_per_2D_tensor ()
 Returns the number of arguments enqueued per 2D tensor object. More...
 
static constexpr unsigned int num_arguments_per_3D_tensor ()
 Returns the number of arguments enqueued per 3D tensor object. More...
 
static constexpr unsigned int num_arguments_per_4D_tensor ()
 Returns the number of arguments enqueued per 4D tensor object. More...
 
static cl::NDRange gws_from_window (const Window &window)
 Get the global work size given an execution window. More...
 

Detailed Description

Interface for the normalization layer kernel.

Definition at line 35 of file CLNormalizationLayerKernel.h.

Constructor & Destructor Documentation

◆ CLNormalizationLayerKernel() [1/3]

Constructor.

Definition at line 124 of file CLNormalizationLayerKernel.cpp.

References arm_compute::ELEMENTWISE.

125  : _input(nullptr), _output(nullptr), _border_size(0), _is_norm_across_width(false)
126 {
128 }
Elementwise CL kernel type.
Definition: CLTypes.h:85

◆ CLNormalizationLayerKernel() [2/3]

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

◆ CLNormalizationLayerKernel() [3/3]

Default Move Constructor.

Member Function Documentation

◆ border_size()

BorderSize border_size ( ) const
overridevirtual

The size of the border for that kernel.

Returns
The width in number of elements of the border.

Reimplemented from IKernel.

Definition at line 130 of file CLNormalizationLayerKernel.cpp.

131 {
132  return _border_size;
133 }

◆ configure() [1/2]

void configure ( const ICLTensor input,
ICLTensor output,
NormalizationLayerInfo  norm_info 
)

Set the input and output tensors.

Parameters
[in]inputSource tensor. 3 lower dims represent a single input with dimensions [width, height, IFM], and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layouts supported: NCHW/NHWC.
[out]outputDestination tensor. Output will have the same number of dimensions as input. Data types supported: same as input. Data layouts supported: same as input.
[in]norm_infoNormalization layer information like the normalization type, normalization size and other parameters.

Definition at line 135 of file CLNormalizationLayerKernel.cpp.

References CLKernelLibrary::get().

136 {
137  configure(CLKernelLibrary::get().get_compile_context(), input, output, norm_info);
138 }
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
void configure(const ICLTensor *input, ICLTensor *output, NormalizationLayerInfo norm_info)
Set the input and output tensors.

◆ configure() [2/2]

void configure ( const CLCompileContext compile_context,
const ICLTensor input,
ICLTensor output,
NormalizationLayerInfo  norm_info 
)

Set the input and output tensors.

Parameters
[in]compile_contextThe compile context to be used.
[in]inputSource tensor. 3 lower dims represent a single input with dimensions [width, height, IFM], and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layouts supported: NCHW/NHWC.
[out]outputDestination tensor. Output will have the same number of dimensions as input. Data types supported: same as input. Data layouts supported: same as input.
[in]norm_infoNormalization layer information like the normalization type, normalization size and other parameters.

Definition at line 140 of file CLNormalizationLayerKernel.cpp.

References CLBuildOptions::add_option(), CLBuildOptions::add_option_if(), arm_compute::adjust_vec_size(), ARM_COMPUTE_ERROR_ON, ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, NormalizationLayerInfo::beta(), arm_compute::create_kernel(), ITensorInfo::data_layout(), ITensorInfo::data_type(), ITensorInfo::dimension(), ITensorInfo::element_size(), arm_compute::float_to_string_with_full_precision(), arm_compute::get_cl_type_from_data_type(), arm_compute::get_normalization_dimension_index(), arm_compute::get_padding_info(), arm_compute::has_padding_changed(), arm_compute::IN_MAP_2D, ITensor::info(), arm_compute::test::validation::input, NormalizationLayerInfo::is_cross_map(), NormalizationLayerInfo::is_in_map(), NormalizationLayerInfo::kappa(), kernel_name, arm_compute::lower_string(), arm_compute::NCHW, arm_compute::NHWC, NormalizationLayerInfo::norm_size(), CLBuildOptions::options(), NormalizationLayerInfo::scale_coeff(), arm_compute::string_from_data_layout(), arm_compute::string_from_data_type(), arm_compute::support::cpp11::to_string(), type, NormalizationLayerInfo::type(), arm_compute::cpu::kernels::validate_and_configure_window(), and arm_compute::cpu::kernels::validate_arguments().

141 {
143  auto padding_info = get_padding_info({ input, output });
144 
145  // Perform validation step
146  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(), output->info(), norm_info));
147  auto win_config = validate_and_configure_window(input->info(), output->info(), norm_info);
148  ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
149 
150  _input = input;
151  _output = output;
152 
153  const DataLayout data_layout = input->info()->data_layout();
154  unsigned int vec_size_x = adjust_vec_size(max_cl_vector_width / input->info()->element_size(), input->info()->dimension(0));
155  int vec_size_x_leftovers = input->info()->dimension(0) % vec_size_x;
156  if(norm_info.is_cross_map() && data_layout == DataLayout::NHWC)
157  {
158  vec_size_x = 1;
159  vec_size_x_leftovers = 0;
160  }
161 
162  if(data_layout == DataLayout::NCHW)
163  {
164  const unsigned int norm_idx = get_normalization_dimension_index(data_layout, norm_info);
165  _is_norm_across_width = norm_idx == 0;
166  const unsigned int norm_radius = norm_info.norm_size() / 2;
167  // Border / padding calculation:
168  // For NCHW no border handling is impelmeneted in the kernel in the x axis.
169  // This means the x axis is fully-padded depending on vec_size_x and norm_size
170  // E.G. for input x dimension = 3, norm_size = 3 (radius = 1), vec_size_x = 2 ('#' is element 'p' is padding):
171  // In : |p|#|#|#|p|p|
172  // Out: |#|#|#|p|
173  // The output has 1 right padding because of the vec_size_x.
174  // The input has 1 left padding because radius = 1.
175  // The input has 2 right padding because of radius = 1 AND the extra output padding
176  const unsigned int border_width_left = _is_norm_across_width ? norm_radius : 0;
177  const unsigned int border_width_right = _is_norm_across_width ? norm_radius + (vec_size_x - input->info()->dimension(0) % vec_size_x) : 0;
178  _border_size = BorderSize(0, border_width_right, 0, border_width_left);
179  }
180 
181  const bool is_in_map_2D = (norm_info.type() == NormType::IN_MAP_2D);
182 
183  // Set build options
184  CLBuildOptions build_opts;
185  build_opts.add_option(("-DDATA_TYPE=" + get_cl_type_from_data_type(input->info()->data_type())));
186  build_opts.add_option(("-DCOEFF=" + float_to_string_with_full_precision(norm_info.scale_coeff())));
187  build_opts.add_option(("-DBETA=" + float_to_string_with_full_precision(norm_info.beta())));
188  build_opts.add_option(("-DKAPPA=" + float_to_string_with_full_precision(norm_info.kappa())));
189  build_opts.add_option(("-DVEC_SIZE=" + support::cpp11::to_string(vec_size_x)));
190  build_opts.add_option(("-DVEC_SIZE_LEFTOVER=" + support::cpp11::to_string(vec_size_x_leftovers)));
191  build_opts.add_option(("-DRADIUS=" + support::cpp11::to_string(norm_info.norm_size() / 2)));
192  build_opts.add_option(("-DNUM_SLICES=" + support::cpp11::to_string(input->info()->dimension(2))));
193  build_opts.add_option_if(is_in_map_2D, "-DIN_MAP_2D");
194  build_opts.add_option_if(norm_info.is_in_map() || (data_layout == DataLayout::NHWC && norm_info.is_cross_map()), "-DWIDTH_SIZE=" + support::cpp11::to_string(input->info()->dimension(0)));
195  build_opts.add_option_if(norm_info.is_in_map() && data_layout == DataLayout::NHWC, "-DDIM1_SIZE=" + support::cpp11::to_string(input->info()->dimension(1)));
196 
197  // Create kernel
198  std::string kernel_name;
199  if(norm_info.is_in_map())
200  {
201  kernel_name = "normalization_layer_in_map_" + lower_string(string_from_data_layout(data_layout));
202  }
203  else
204  {
205  kernel_name = "normalization_layer_cross_map_" + lower_string(string_from_data_layout(data_layout));
206  }
207  _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
208 
209  // Configure kernel window
210  ICLKernel::configure_internal(win_config.second);
211 
212  // Set config_id for enabling LWS tuning
213  _config_id = "normalization_layer_";
214  _config_id += lower_string(string_from_data_type(input->info()->data_type()));
215  _config_id += "_";
216  _config_id += support::cpp11::to_string(static_cast<std::underlying_type<NormType>::type>(norm_info.type()));
217  _config_id += "_";
218  _config_id += support::cpp11::to_string(norm_info.norm_size());
219  _config_id += "_";
220  _config_id += support::cpp11::to_string(input->info()->dimension(0));
221  _config_id += "_";
222  _config_id += support::cpp11::to_string(input->info()->dimension(1));
223  if(data_layout == DataLayout::NHWC)
224  {
226  }
227 }
std::string to_string(T &&value)
Convert integer and float values to string.
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
Definition: Error.h:466
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
std::string lower_string(const std::string &val)
Lower a given string.
Definition: Utils.cpp:353
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
decltype(strategy::transforms) typedef type
cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set< std::string > &build_opts=std::set< std::string >())
Creates an opencl kernel using a compile context.
Definition: CLHelpers.cpp:404
const std::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
Definition: Utils.cpp:135
std::string float_to_string_with_full_precision(float val)
Create a string with the float in full precision.
Definition: Utils.h:1124
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
Definition: CLHelpers.cpp:39
bool has_padding_changed(const std::unordered_map< const ITensorInfo *, PaddingSize > &padding_map)
Check if the previously stored padding info has changed after configuring a kernel.
Definition: Utils.cpp:603
Num samples, channels, height, width.
std::pair< Status, Window > validate_and_configure_window(ITensorInfo *src, ITensorInfo *dst)
const std::string & string_from_data_layout(DataLayout dl)
Convert a data layout identity into a string.
Definition: Utils.cpp:123
Num samples, height, width, channels.
std::unordered_map< const ITensorInfo *, PaddingSize > get_padding_info(std::initializer_list< const ITensorInfo *> infos)
Stores padding information before configuring a kernel.
Definition: Utils.cpp:588
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:157
unsigned int adjust_vec_size(unsigned int vec_size, size_t dim0)
Returns the adjusted vector size in case it is less than the input&#39;s first dimension, getting rounded down to its closest valid vector size.
Definition: Utils.h:1222
std::string kernel_name
unsigned int get_normalization_dimension_index(DataLayout layout, const NormalizationLayerInfo &info)
Calculate the normalization dimension index for a given normalization type.
DataLayout
[DataLayout enum definition]
Definition: Types.h:113
Normalization applied within the same map in 2D region.

◆ operator=() [1/2]

CLNormalizationLayerKernel& operator= ( const CLNormalizationLayerKernel )
delete

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

◆ operator=() [2/2]

Default move assignment operator.

◆ run()

void run ( const Window window,
cl::CommandQueue &  queue 
)
overridevirtual

Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue.

Note
The queue is not flushed by this method, and therefore the kernel will not have been executed by the time this method returns.
Parameters
[in]windowRegion on which to execute the kernel. (Must be a valid region of the window returned by window()).
[in,out]queueCommand queue on which to enqueue the kernel.

Reimplemented from ICLKernel.

Definition at line 237 of file CLNormalizationLayerKernel.cpp.

References ICLKernel::add_3D_tensor_argument(), ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, Window::collapse_if_possible(), Window::DimZ, arm_compute::enqueue(), Window::first_slice_window_3D(), ICLKernel::lws_hint(), arm_compute::test::validation::reference::slice(), Window::slide_window_slice_3D(), and IKernel::window().

238 {
241 
242  const int collapsed_dimension = _is_norm_across_width ? Window::DimZ : 4;
243  Window window_collapsed = window.collapse_if_possible(ICLKernel::window(), collapsed_dimension);
244  Window slice = window_collapsed.first_slice_window_3D();
245 
246  do
247  {
248  unsigned int idx = 0;
249  add_3D_tensor_argument(idx, _input, slice);
250  add_3D_tensor_argument(idx, _output, slice);
251  enqueue(queue, *this, slice, lws_hint());
252  }
253  while(window_collapsed.slide_window_slice_3D(slice));
254 }
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
void enqueue(cl::CommandQueue &queue, ICLKernel &kernel, const Window &window, const cl::NDRange &lws_hint=CLKernelLibrary::get().default_ndrange(), bool use_dummy_work_items=false)
Add the kernel to the command queue with the given window.
Definition: ICLKernel.cpp:32
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
Definition: ICLKernel.h:383
void add_3D_tensor_argument(unsigned int &idx, const ICLTensor *tensor, const Window &window)
Add the passed 3D tensor&#39;s parameters to the object&#39;s kernel&#39;s arguments starting from the index idx...
Definition: ICLKernel.h:226
Window collapse_if_possible(const Window &full_window, size_t first, size_t last, bool *has_collapsed=nullptr) const
Collapse the dimensions between first and last if possible.
Definition: Window.inl:68
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:915
static constexpr size_t DimZ
Alias for dimension 2 also known as Z dimension.
Definition: Window.h:47
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:201
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)

◆ validate()

Status validate ( const ITensorInfo input,
const ITensorInfo output,
NormalizationLayerInfo  norm_info 
)
static

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

Parameters
[in]inputSource tensor. 3 lower dims represent a single input with dimensions [width, height, IFM], and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layouts supported: NCHW/NHWC.
[in]outputDestination tensor. Output will have the same number of dimensions as input. Data types supported: same as input. Data layouts supported: same as input.
[in]norm_infoNormalization layer information like the normalization type, normalization size and other parameters.
Returns
a status

Definition at line 229 of file CLNormalizationLayerKernel.cpp.

References ARM_COMPUTE_RETURN_ON_ERROR, ICloneable< T >::clone(), arm_compute::cpu::kernels::validate_and_configure_window(), and arm_compute::cpu::kernels::validate_arguments().

Referenced by CLNormalizationLayer::validate().

230 {
232  ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(input->clone().get(), output->clone().get(), norm_info).first);
233 
234  return Status{};
235 }
#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 *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
std::pair< Status, Window > validate_and_configure_window(ITensorInfo *src, ITensorInfo *dst)

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