Compute Library
 22.05
ClActivationKernel Class Reference

Interface for the activation kernel. More...

#include <ClActivationKernel.h>

Collaboration diagram for ClActivationKernel:
[legend]

Public Member Functions

 ClActivationKernel ()
 
 ARM_COMPUTE_DISALLOW_COPY_ALLOW_MOVE (ClActivationKernel)
 
void configure (const ClCompileContext &compile_context, ITensorInfo *src, ITensorInfo *dst, ActivationLayerInfo act_info)
 Configure kernel for a given list of arguments. More...
 
void run_op (ITensorPack &tensors, const Window &window, ::cl::CommandQueue &queue) override
 
- 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 (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_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...
 
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...
 
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 *src, const ITensorInfo *dst, const ActivationLayerInfo &act_info)
 Static function to check if given info will lead to a valid configuration. 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 activation kernel.

Definition at line 38 of file ClActivationKernel.h.

Constructor & Destructor Documentation

◆ ClActivationKernel()

Definition at line 89 of file ClActivationKernel.cpp.

References arm_compute::ELEMENTWISE.

90 {
92 }
Elementeise CL kernel type.
Definition: CLTypes.h:84

Member Function Documentation

◆ ARM_COMPUTE_DISALLOW_COPY_ALLOW_MOVE()

ARM_COMPUTE_DISALLOW_COPY_ALLOW_MOVE ( ClActivationKernel  )

◆ configure()

void configure ( const ClCompileContext compile_context,
ITensorInfo src,
ITensorInfo dst,
ActivationLayerInfo  act_info 
)

Configure kernel for a given list of arguments.

Note
If the output tensor is a nullptr, the activation function will be performed in-place
Parameters
[in]compile_contextThe compile context to be used.
[in,out]srcSource tensor info. In case of dst tensor = nullptr, this tensor will store the result of the activation function. Data types supported: QASYMM8/QASYMM8_SIGNED/QSYMM16/F16/F32.
[out]dstDestination tensor info. Data type supported: same as src
[in]act_infoActivation layer information.

Definition at line 94 of file ClActivationKernel.cpp.

References ActivationLayerInfo::a(), ActivationLayerInfo::activation(), 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, arm_compute::auto_init_if_empty(), ActivationLayerInfo::b(), arm_compute::calculate_max_window(), ICloneable< T >::clone(), arm_compute::create_kernel(), ITensorInfo::data_type(), ITensorInfo::dimension(), dt, ITensorInfo::element_size(), arm_compute::float_to_string_with_full_precision(), arm_compute::get_cl_type_from_data_type(), arm_compute::get_padding_info(), ActivationLayerInfo::HARD_SWISH, arm_compute::has_padding_changed(), arm_compute::is_data_type_quantized(), arm_compute::is_data_type_quantized_asymmetric(), kernel_name, ActivationLayerInfo::LEAKY_RELU, ActivationLayerInfo::LOGISTIC, arm_compute::lower_string(), num_elems_processed_per_iteration, UniformQuantizationInfo::offset, CLBuildOptions::options(), arm_compute::QASYMM8, arm_compute::QASYMM8_SIGNED, arm_compute::QSYMM16, ITensorInfo::quantization_info(), arm_compute::quantize_qasymm8(), arm_compute::quantize_qasymm8_signed(), arm_compute::quantize_qsymm16(), UniformQuantizationInfo::scale, arm_compute::test::validation::src, arm_compute::string_from_activation_func(), arm_compute::string_from_data_type(), ActivationLayerInfo::TANH, arm_compute::support::cpp11::to_string(), QuantizationInfo::uniform(), and arm_compute::cpu::kernels::validate_arguments().

95 {
97 
98  auto padding_info = get_padding_info({ src, dst });
99 
100  _run_in_place = (dst == nullptr) || (dst == src);
101 
102  if(dst != nullptr)
103  {
104  // Destination auto inizialitation if not yet initialized
105  auto_init_if_empty(*dst, *src->clone());
106  }
107 
108  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src, (dst != nullptr) ? dst : nullptr, act_info));
109 
110  const unsigned int num_elems_processed_per_iteration = adjust_vec_size(16 / src->element_size(), src->dimension(0));
111 
112  const DataType dt = src->data_type();
113  float a_const = act_info.a();
114  float b_const = act_info.b();
115 
116  const ActivationLayerInfo::ActivationFunction f_act = act_info.activation();
117  const bool is_quantized = is_data_type_quantized(dt);
118  const bool perform_activation_in_float =
123 
124  // Set build options
125  CLBuildOptions build_opts;
126  build_opts.add_option_if(perform_activation_in_float, "-DFLOAT_DOMAIN");
127  build_opts.add_option_if(_run_in_place, "-DIN_PLACE");
128  build_opts.add_option("-DACT=" + lower_string(string_from_activation_func(f_act)));
129  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(dt));
130  build_opts.add_option("-DVEC_SIZE=" + support::cpp11::to_string(num_elems_processed_per_iteration));
131  build_opts.add_option("-DVEC_SIZE_LEFTOVER=" + support::cpp11::to_string(src->dimension(0) % num_elems_processed_per_iteration));
132 
133  std::string kernel_name = std::string("activation_layer");
134 
135  // Set quantization info build options
136  if(is_quantized)
137  {
138  const UniformQuantizationInfo iq_info = src->quantization_info().uniform();
139 
140  if(!perform_activation_in_float)
141  {
142  int a_const_int = 0;
143  int b_const_int = 0;
144 
145  // Create quantized version of constants a, b if needed
146  switch(dt)
147  {
148  case DataType::QASYMM8:
149  {
150  a_const_int = quantize_qasymm8(a_const, iq_info);
151  b_const_int = quantize_qasymm8(b_const, iq_info);
152  }
153  break;
155  {
156  a_const_int = quantize_qasymm8_signed(a_const, iq_info);
157  b_const_int = quantize_qasymm8_signed(b_const, iq_info);
158  }
159  break;
160  case DataType::QSYMM16:
161  {
162  a_const_int = quantize_qsymm16(a_const, iq_info);
163  b_const_int = quantize_qsymm16(b_const, iq_info);
164  }
165  break;
166  default:
167  break;
168  }
169  build_opts.add_option(("-DA_VAL=" + support::cpp11::to_string(a_const_int)));
170  build_opts.add_option(("-DB_VAL=" + support::cpp11::to_string(b_const_int)));
171  }
172  else
173  {
174  build_opts.add_option(("-DA_VAL=" + float_to_string_with_full_precision(a_const)));
175  build_opts.add_option(("-DB_VAL=" + float_to_string_with_full_precision(b_const)));
176  }
177 
178  // Quantized value of 0 corresponds to the offset o1
179  build_opts.add_option(("-DCONST_0=" + (is_data_type_quantized_asymmetric(dt) ? support::cpp11::to_string(iq_info.offset) : "0")));
180  build_opts.add_option(("-DS1_VAL=" + float_to_string_with_full_precision(iq_info.scale)));
181  build_opts.add_option_if(is_data_type_quantized_asymmetric(dt), "-DO1_VAL=" + support::cpp11::to_string(iq_info.offset));
182 
183  // Set correct kernel name
184  kernel_name += perform_activation_in_float ? std::string("_quant_f32") : std::string("_quant");
185 
186  // Set scale and offset of the source and destination if they have different quantization info
187  if(dst != nullptr)
188  {
189  const UniformQuantizationInfo oq_info = dst->quantization_info().uniform();
190 
191  if(iq_info != oq_info)
192  {
193  build_opts.add_option(("-DS2_VAL=" + float_to_string_with_full_precision(oq_info.scale)));
194  build_opts.add_option_if(is_data_type_quantized_asymmetric(dt), "-DO2_VAL=" + support::cpp11::to_string(oq_info.offset));
195  }
196  }
197  }
198  else
199  {
200  // Set A, B constants in build options for float types
201  build_opts.add_option(("-DA_VAL=" + float_to_string_with_full_precision(a_const)));
202  build_opts.add_option(("-DB_VAL=" + float_to_string_with_full_precision(b_const)));
203  }
204 
205  // Create kernel
206  _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
207 
208  // Configure kernel window
209  Window win = calculate_max_window(*src, Steps(num_elems_processed_per_iteration));
210  ICLKernel::configure_internal(win);
211 
212  // Set config_id for enabling LWS tuning
213  _config_id = "activation_layer_";
214  _config_id += lower_string(string_from_data_type(dt));
215  _config_id += "_";
216  _config_id += support::cpp11::to_string(src->dimension(0));
217  _config_id += "_";
218  _config_id += support::cpp11::to_string(src->dimension(1));
219 
221 }
bool is_data_type_quantized(DataType dt)
Check if a given data type is of quantized type.
Definition: Utils.h:1030
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
int16_t quantize_qsymm16(float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
Quantize a value given a 16-bit symmetric quantization scheme.
quantized, symmetric fixed-point 16-bit number
uint8_t quantize_qasymm8(float value, const INFO_TYPE &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
Quantize a value given an unsigned 8-bit asymmetric quantization scheme.
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
const std::string & string_from_activation_func(ActivationLayerInfo::ActivationFunction act)
Translates a given activation function to a string.
Definition: Utils.cpp:163
#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:351
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
SimpleTensor< float > src
Definition: DFT.cpp:155
ActivationFunction
Available activation functions.
Definition: Types.h:1629
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:391
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
int8_t quantize_qasymm8_signed(float value, const INFO_TYPE &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
Quantize a value given a signed 8-bit asymmetric quantization scheme.
quantized, asymmetric fixed-point 8-bit number unsigned
unsigned int num_elems_processed_per_iteration
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 auto_init_if_empty(ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())
Auto initialize the tensor info (shape, number of channels and data type) if the current assignment i...
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:601
bool is_data_type_quantized_asymmetric(DataType dt)
Check if a given data type is of asymmetric quantized type.
Definition: Utils.h:1052
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:586
#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
quantized, asymmetric fixed-point 8-bit number signed
std::string kernel_name
DataType
Available data types.
Definition: Types.h:79

◆ run_op()

void run_op ( ITensorPack tensors,
const Window window,
::cl::CommandQueue &  queue 
)
override

Definition at line 229 of file ClActivationKernel.cpp.

References arm_compute::ACL_DST, arm_compute::ACL_SRC, ICLKernel::add_3D_tensor_argument(), ARM_COMPUTE_ERROR_ON, 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(), ITensorPack::get_const_tensor(), ITensorPack::get_tensor(), ICLKernel::lws_hint(), arm_compute::test::validation::reference::slice(), Window::slide_window_slice_3D(), and IKernel::window().

230 {
233 
234  const auto src = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC));
235  auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
236  ARM_COMPUTE_ERROR_ON(_run_in_place && src != dst);
237 
239  Window slice = collapsed.first_slice_window_3D();
240 
241  do
242  {
243  unsigned int idx = 0;
244  add_3D_tensor_argument(idx, src, slice);
245  if(!_run_in_place)
246  {
247  add_3D_tensor_argument(idx, dst, slice);
248  }
249  enqueue(queue, *this, slice, lws_hint());
250  }
251  while(collapsed.slide_window_slice_3D(slice));
252 }
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:384
#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
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:227
SimpleTensor< float > src
Definition: DFT.cpp:155
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 src,
const ITensorInfo dst,
const ActivationLayerInfo act_info 
)
static

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

Similar to ClActivationKernel::configure()

Returns
a status

Definition at line 223 of file ClActivationKernel.cpp.

References ARM_COMPUTE_RETURN_ON_ERROR, and arm_compute::cpu::kernels::validate_arguments().

Referenced by ClActivation::validate(), ClDirectConv2d::validate(), and ClGemmConv2d::validate().

224 {
226  return Status{};
227 }
#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)
SimpleTensor< float > src
Definition: DFT.cpp:155

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