Compute Library
 20.08
CLQLSTMLayerNormalizationKernel Class Reference

Interface for the kernel to do layer normalization. More...

#include <CLQLSTMLayerNormalizationKernel.h>

Collaboration diagram for CLQLSTMLayerNormalizationKernel:
[legend]

Public Member Functions

 CLQLSTMLayerNormalizationKernel ()
 Default constructor. More...
 
 CLQLSTMLayerNormalizationKernel (const CLQLSTMLayerNormalizationKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
CLQLSTMLayerNormalizationKerneloperator= (const CLQLSTMLayerNormalizationKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 CLQLSTMLayerNormalizationKernel (CLQLSTMLayerNormalizationKernel &&)=default
 Allow instances of this class to be moved. More...
 
CLQLSTMLayerNormalizationKerneloperator= (CLQLSTMLayerNormalizationKernel &&)=default
 Allow instances of this class to be moved. More...
 
 ~CLQLSTMLayerNormalizationKernel ()=default
 Default destructor. More...
 
void configure (const ICLTensor *input, ICLTensor *output, const ICLTensor *weight, const ICLTensor *bias)
 Initialise the kernel's input and outputs. More...
 
void configure (const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *weight, const ICLTensor *bias)
 Initialise the kernel's input and outputs. 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...
 
- Public Member Functions inherited from ICLKernel
 ICLKernel ()
 Constructor. More...
 
cl::Kernel & kernel ()
 Returns a reference to the OpenCL kernel of this object. 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...
 
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...
 
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...
 
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<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...
 
template<unsigned int dimension_size>
void add_tensor_argument (unsigned &idx, const ICLTensor *tensor, const Window &window)
 
- 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 *input, const ITensorInfo *output, const ITensorInfo *weight, const ITensorInfo *bias)
 Static function to check if given info will lead to a valid configuration of CLQLSTMLayerNormalizationKernel. More...
 
- Static Public Member Functions inherited from ICLKernel
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 kernel to do layer normalization.

Definition at line 34 of file CLQLSTMLayerNormalizationKernel.h.

Constructor & Destructor Documentation

◆ CLQLSTMLayerNormalizationKernel() [1/3]

Default constructor.

Definition at line 82 of file CLQLSTMLayerNormalizationKernel.cpp.

83  : _input(nullptr), _weight(nullptr), _bias(nullptr), _output(nullptr)
84 {
85 }

◆ CLQLSTMLayerNormalizationKernel() [2/3]

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

◆ CLQLSTMLayerNormalizationKernel() [3/3]

Allow instances of this class to be moved.

◆ ~CLQLSTMLayerNormalizationKernel()

Default destructor.

Member Function Documentation

◆ configure() [1/2]

void configure ( const ICLTensor input,
ICLTensor output,
const ICLTensor weight,
const ICLTensor bias 
)

Initialise the kernel's input and outputs.

Parameters
[in]inputSource tensor with 2 dimensions. Data types supported: QSYMM16.
[out]outputDestination tensor. Data type supported: same as input
[in]weightWeight tensor. Data types supported: Same as input.
[in]biasBias tensor. Data types supported: S32.

Definition at line 133 of file CLQLSTMLayerNormalizationKernel.cpp.

134 {
135  configure(CLKernelLibrary::get().get_compile_context(), input, output, weight, bias);
136 }
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
void configure(const ICLTensor *input, ICLTensor *output, const ICLTensor *weight, const ICLTensor *bias)
Initialise the kernel's input and outputs.

References arm_compute::test::validation::bias, CLKernelLibrary::get(), and arm_compute::test::validation::input.

◆ configure() [2/2]

void configure ( const CLCompileContext compile_context,
const ICLTensor input,
ICLTensor output,
const ICLTensor weight,
const ICLTensor bias 
)

Initialise the kernel's input and outputs.

Parameters
[in]compile_contextThe compile context to be used.
[in]inputSource tensor with 2 dimensions. Data types supported: QSYMM16.
[out]outputDestination tensor. Data type supported: same as input
[in]weightWeight tensor. Data types supported: Same as input.
[in]biasBias tensor. Data types supported: S32.

Definition at line 87 of file CLQLSTMLayerNormalizationKernel.cpp.

88 {
89  ARM_COMPUTE_ERROR_ON_NULLPTR(input, weight, bias, output);
90 
91  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(), output->info(), weight->info(), bias->info()));
92 
93  _input = input;
94  _weight = weight;
95  _bias = bias;
96  _output = output;
97 
98  const uint32_t num_elems_processed_per_iteration = max_cl_vector_width / input->info()->element_size();
99 
100  int32_t output_multiplier{};
101  int32_t output_shift{};
102  const UniformQuantizationInfo quan_info = _weight->info()->quantization_info().uniform();
103  const Status status = quantization::calculate_quantized_multiplier(quan_info.scale, &output_multiplier, &output_shift);
104  output_shift *= -1;
105 
106  // Set build options
107  CLBuildOptions build_opts;
108  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(input->info()->data_type()));
109  build_opts.add_option("-DVEC_SIZE=" + support::cpp11::to_string(num_elems_processed_per_iteration));
110  build_opts.add_option("-DWIDTH=" + support::cpp11::to_string(input->info()->dimension(0)));
111  build_opts.add_option("-DOUTPUT_MULTIPLIER=" + support::cpp11::to_string(output_multiplier));
112  build_opts.add_option("-DOUTPUT_SHIFT=" + support::cpp11::to_string(output_shift));
113  build_opts.add_option("-DMIN_BOUND=" + support::cpp11::to_string(std::get<0>(quantization::get_min_max_values_from_quantized_data_type(input->info()->data_type()))));
114  build_opts.add_option("-DMAX_BOUND=" + support::cpp11::to_string(std::get<1>(quantization::get_min_max_values_from_quantized_data_type(input->info()->data_type()))));
115 
116  // Create kernel
117  _kernel = create_kernel(compile_context, "qlstm_layer_normalization", build_opts.options());
118 
119  // Configure kernel window
120  auto win_config = validate_and_configure_window(input->info(), output->info());
121  ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
122  ICLKernel::configure_internal(win_config.second);
123 
124  // Set config_id for enabling LWS tuning
125  _config_id = "qlstm_layer_normalization_";
126  _config_id += lower_string(string_from_data_type(input->info()->data_type()));
127  _config_id += "_";
128  _config_id += support::cpp11::to_string(input->info()->dimension(0));
129  _config_id += "_";
130  _config_id += support::cpp11::to_string(input->info()->dimension(1));
131 }
std::string to_string(T &&value)
Convert integer and float values to string.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
Status calculate_quantized_multiplier(float multiplier, int32_t *quant_multiplier, int32_t *shift, bool ignore_epsilon=false)
Calculate quantized representation of multiplier.
std::string lower_string(const std::string &val)
Lower a given string.
Definition: Utils.cpp:326
ITensorInfo * info() const override
Interface to be implemented by the child class to return the tensor's metadata.
Definition: Tensor.cpp:33
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:403
const std::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
Definition: Utils.cpp:135
std::pair< int, int > get_min_max_values_from_quantized_data_type(DataType data_type)
Get minimum and maximum values for the input quantized data type.
UniformQuantizationInfo uniform() const
Return per layer quantization info.
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
Definition: CLHelpers.cpp:37
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
virtual QuantizationInfo quantization_info() const =0
Get the quantization settings (scale and offset) of the tensor.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161
Status validate_arguments(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo *output_stage)
unsigned int num_elems_processed_per_iteration

References CLBuildOptions::add_option(), ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, arm_compute::test::validation::bias, arm_compute::quantization::calculate_quantized_multiplier(), arm_compute::create_kernel(), arm_compute::get_cl_type_from_data_type(), arm_compute::quantization::get_min_max_values_from_quantized_data_type(), ITensor::info(), Tensor::info(), arm_compute::test::validation::input, arm_compute::lower_string(), num_elems_processed_per_iteration, CLBuildOptions::options(), ITensorInfo::quantization_info(), UniformQuantizationInfo::scale, arm_compute::string_from_data_type(), arm_compute::support::cpp11::to_string(), QuantizationInfo::uniform(), and arm_compute::validate_arguments().

◆ operator=() [1/2]

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

◆ operator=() [2/2]

Allow instances of this class to be moved.

◆ 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 145 of file CLQLSTMLayerNormalizationKernel.cpp.

146 {
149 
151  // Set slice step equal to width to force gws[0] to 1, as each thread normalizes across all rows
152  slice.set_dimension_step(Window::DimX, _input->info()->dimension(0));
153 
154  Window weight_window;
155  Window weight_slice;
156 
157  weight_window.use_tensor_dimensions(_weight->info()->tensor_shape());
158  weight_slice = weight_window.first_slice_window_1D();
159 
160  do
161  {
162  unsigned int idx = 0;
163  add_2D_tensor_argument(idx, _input, slice);
164  add_1D_tensor_argument(idx, _weight, weight_slice);
165  add_1D_tensor_argument(idx, _bias, weight_slice);
166  add_2D_tensor_argument(idx, _output, slice);
167 
168  enqueue(queue, *this, slice, lws_hint());
169  }
171 }
Window first_slice_window_2D() const
First 2D slice of the window.
Definition: Window.h:281
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
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:39
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
Definition: ICLKernel.h:263
bool slide_window_slice_2D(Window &slice) const
Slide the passed 2D window slice.
Definition: Window.h:321
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Definition: Window.h:43
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's metadata.
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.
Definition: ICLKernel.h:135
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.
Definition: ICLKernel.h:111
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:205
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:941
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)

References ICLKernel::add_1D_tensor_argument(), ICLKernel::add_2D_tensor_argument(), ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, ITensorInfo::dimension(), Window::DimX, arm_compute::enqueue(), Window::first_slice_window_1D(), Window::first_slice_window_2D(), ITensor::info(), ICLKernel::lws_hint(), arm_compute::test::validation::reference::slice(), Window::slide_window_slice_2D(), ITensorInfo::tensor_shape(), Window::use_tensor_dimensions(), and IKernel::window().

◆ validate()

Status validate ( const ITensorInfo input,
const ITensorInfo output,
const ITensorInfo weight,
const ITensorInfo bias 
)
static

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

Parameters
[in]inputSource tensor info with 2 dimensions. Data types supported: QSYMM16.
[in]outputDestination info tensor. Data type supported: same as input
[in]weightWeight info tensor. Data types supported: Same as input.
[in]biasBias tensor info. Data types supported: S32.
Returns
a status

Definition at line 138 of file CLQLSTMLayerNormalizationKernel.cpp.

139 {
141  ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(input->clone().get(), output->clone().get()).first);
142  return Status{};
143 }
#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)

References ARM_COMPUTE_RETURN_ON_ERROR, arm_compute::test::validation::bias, ICloneable< T >::clone(), arm_compute::test::validation::input, and arm_compute::validate_arguments().


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