Compute Library
 19.08
CLStridedSliceKernel Class Reference

Interface for the kernel to perform tensor strided slicing. More...

#include <CLStridedSliceKernel.h>

Collaboration diagram for CLStridedSliceKernel:
[legend]

Public Member Functions

 CLStridedSliceKernel ()
 Default constructor. More...
 
 CLStridedSliceKernel (const CLStridedSliceKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
CLStridedSliceKerneloperator= (const CLStridedSliceKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 CLStridedSliceKernel (CLStridedSliceKernel &&)=default
 Allow instances of this class to be moved. More...
 
CLStridedSliceKerneloperator= (CLStridedSliceKernel &&)=default
 Allow instances of this class to be moved. More...
 
 ~CLStridedSliceKernel ()=default
 Default destructor. More...
 
void configure (const ICLTensor *input, ICLTensor *output, const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, int32_t begin_mask, int32_t end_mask, int32_t shrink_axis_mask)
 Configure kernel. 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...
 
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 Coordinates &starts, const Coordinates &ends, const BiStrides &strides, int32_t begin_mask, int32_t end_mask, int32_t shrink_axis_mask)
 Static function to check if given info will lead to a valid configuration of CLStridedSliceKernel. 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 perform tensor strided slicing.

Definition at line 38 of file CLStridedSliceKernel.h.

Constructor & Destructor Documentation

◆ CLStridedSliceKernel() [1/3]

Default constructor.

Definition at line 99 of file CLStridedSliceKernel.cpp.

100  : _input(nullptr), _output(nullptr)
101 {
102 }

◆ CLStridedSliceKernel() [2/3]

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

◆ CLStridedSliceKernel() [3/3]

Allow instances of this class to be moved.

◆ ~CLStridedSliceKernel()

~CLStridedSliceKernel ( )
default

Default destructor.

Member Function Documentation

◆ configure()

void configure ( const ICLTensor input,
ICLTensor output,
const Coordinates starts,
const Coordinates ends,
const BiStrides strides,
int32_t  begin_mask,
int32_t  end_mask,
int32_t  shrink_axis_mask 
)

Configure kernel.

Note
Supported tensor rank: up to 4
Parameters
[in]inputSource tensor. Data type supported: U8/S8/QASYMM8/U16/S16/QSYMM16/U32/S32/F16/F32
[out]outputDestination tensor. Data type supported: Same as input
[in]startsThe starts of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]endsThe ends of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]stridesThe strides of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]begin_maskIf the ith bit of begin_mask is set, starts[i] is ignored and the fullest possible range in that dimension is used instead.
[in]end_maskIf the ith bit of end_mask is set, ends[i] is ignored and the fullest possible range in that dimension is used instead.
[in]shrink_axis_maskIf the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. A slice of size 1 starting from starts[i] in the dimension must be preserved.

Definition at line 104 of file CLStridedSliceKernel.cpp.

107 {
108  ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
109  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(), output->info(), starts, ends, strides, begin_mask, end_mask, shrink_axis_mask));
110 
111  _input = input;
112  _output = output;
113 
114  const TensorShape &input_shape = input->info()->tensor_shape();
115 
116  Coordinates starts_abs;
117  Coordinates ends_abs;
118  Coordinates final_strides;
119  std::tie(starts_abs, ends_abs, final_strides) = arm_compute::helpers::tensor_transform::calculate_strided_slice_coords(
120  input_shape,
121  starts, ends, strides,
122  begin_mask, end_mask, shrink_axis_mask);
123 
124  // Configure kernel window
125  auto win_config = validate_and_configure_window(input->info(), output->info(), starts, ends, strides, begin_mask, end_mask, shrink_axis_mask);
126  ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
127 
128  // Enable multiple elements processing along x if stride_x is 1 and output width greater than the access vector size
129  const int vec_size_x = 16 / input->info()->element_size();
130  const int output_width_x = output->info()->tensor_shape().x();
131  const bool is_shrink_on_x = arm_compute::helpers::bit_ops::is_bit_set(shrink_axis_mask, 0);
132  const bool multi_access_x = !is_shrink_on_x && (final_strides.x() == 1) && (output_width_x / vec_size_x > 0);
133 
134  // Update window if needed
135  if(multi_access_x)
136  {
137  Window &updated_window = std::get<1>(win_config);
138  updated_window.set(Window::DimX,
139  Window::Dimension(updated_window.x().start(), ceil_to_multiple(updated_window.x().end(), vec_size_x), vec_size_x));
140  }
141  ICLKernel::configure_internal(win_config.second);
142 
143  // Create build options
144  CLBuildOptions build_opts;
145  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(input->info()->data_type()));
146  for(unsigned int i = 0; i < input_shape.num_dimensions(); ++i)
147  {
148  const bool is_shrink = arm_compute::helpers::bit_ops::is_bit_set(shrink_axis_mask, i);
149  build_opts.add_option("-DSTART_" + support::cpp11::to_string(i) + "=" + support::cpp11::to_string(starts_abs[i]));
150  build_opts.add_option("-DSTRIDE_" + support::cpp11::to_string(i) + "=" + support::cpp11::to_string(final_strides[i]));
151  build_opts.add_option_if(is_shrink, "-DSHRINK_" + support::cpp11::to_string(i));
152  }
153  build_opts.add_option_if(multi_access_x, "-DLAST_ACCESSED_X=" + support::cpp11::to_string(std::max<int>(output_width_x - vec_size_x, 0)));
154  build_opts.add_option_if(multi_access_x, "-DVEC_SIZE=" + support::cpp11::to_string(vec_size_x));
155  build_opts.add_option_if_else(input_shape.num_dimensions() > 2,
156  "-DSRC_DEPTH=" + support::cpp11::to_string(input_shape.z()),
157  "-DSRC_DEPTH=1");
158  build_opts.add_option_if_else(_output->info()->num_dimensions() > 2,
159  "-DDST_DEPTH=" + support::cpp11::to_string(_output->info()->tensor_shape().z()),
160  "-DDST_DEPTH=1");
161 
162  // Create kernel
163  _kernel = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel("strided_slice", build_opts.options()));
164 
165  // Set config_id for enabling LWS tuning
166  _config_id = "strided_slice";
167  _config_id += "_";
168  _config_id += lower_string(string_from_data_type(input->info()->data_type()));
169  for(unsigned int i = 0; i < input_shape.num_dimensions(); ++i)
170  {
171  _config_id += "_";
172  _config_id += support::cpp11::to_string(input->info()->dimension(i));
173  _config_id += "_";
174  _config_id += support::cpp11::to_string(starts_abs[i]);
175  _config_id += "_";
176  _config_id += support::cpp11::to_string(ends_abs[i]);
177  _config_id += "_";
178  _config_id += support::cpp11::to_string(final_strides[i]);
179  }
180 }
virtual size_t num_dimensions() const =0
The number of dimensions of the tensor (rank)
std::tuple< Coordinates, Coordinates, Coordinates > calculate_strided_slice_coords(TensorShape input_shape, Coordinates starts, Coordinates ends, Coordinates strides, int32_t begin_mask=0, int32_t end_mask=0, int32_t shrink_axis_mask=0)
Calculate start, end and stride coordinates for a strided slice.
std::pair< Status, Window > validate_and_configure_window(ITensorInfo *input, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier, const Size2D &dilation)
bool is_bit_set(T v, unsigned int idx)
Checks if the idx-th bit is set in an integral type.
Definition: bit_ops.h:45
std::string to_string(T &&value)
Convert integer and float values to string.
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:327
std::string lower_string(const std::string &val)
Lower a given string.
Definition: Utils.cpp:327
const std::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
Definition: Utils.cpp:144
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.
auto ceil_to_multiple(S value, T divisor) -> decltype(((value+divisor - 1)/divisor) *divisor)
Computes the smallest number larger or equal to value that is a multiple of divisor.
Definition: Utils.h:66
T z() const
Alias to access the size of the third dimension.
Definition: Dimensions.h:91
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
Definition: CLHelpers.cpp:35
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
std::unique_ptr< Kernel > create_kernel()
Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object.
Definition: Helpers.h:86
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161

References CLBuildOptions::add_option(), CLBuildOptions::add_option_if(), CLBuildOptions::add_option_if_else(), ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, arm_compute::helpers::tensor_transform::calculate_strided_slice_coords(), arm_compute::ceil_to_multiple(), arm_compute::create_kernel(), ITensorInfo::data_type(), ITensorInfo::dimension(), Window::DimX, ITensorInfo::element_size(), Window::Dimension::end(), CLKernelLibrary::get(), arm_compute::get_cl_type_from_data_type(), ITensor::info(), arm_compute::test::validation::input_shape, arm_compute::helpers::bit_ops::is_bit_set(), arm_compute::lower_string(), ITensorInfo::num_dimensions(), CLBuildOptions::options(), Window::set(), Window::Dimension::start(), arm_compute::string_from_data_type(), ITensorInfo::tensor_shape(), arm_compute::support::cpp11::to_string(), arm_compute::validate_and_configure_window(), Dimensions< T >::x(), Window::x(), and Dimensions< T >::z().

◆ operator=() [1/2]

CLStridedSliceKernel& operator= ( const CLStridedSliceKernel )
delete

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

◆ operator=() [2/2]

CLStridedSliceKernel& operator= ( CLStridedSliceKernel &&  )
default

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.

Implements ICLKernel.

Definition at line 194 of file CLStridedSliceKernel.cpp.

195 {
198 
199  Window window_collapsed = window.collapse_if_possible(ICLKernel::window(), Window::DimZ);
200  Window slice = window_collapsed.first_slice_window_4D();
201 
202  do
203  {
204  unsigned int idx = 0;
205  add_4D_tensor_argument(idx, _input, slice);
206  add_4D_tensor_argument(idx, _output, slice);
207  enqueue(queue, *this, slice, lws_hint());
208  }
209  while(window_collapsed.slide_window_slice_4D(slice));
210 }
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:39
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
Definition: ICLKernel.h:247
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:54
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:205
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.
Definition: ICLKernel.h:168
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:940
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)

References ICLKernel::add_4D_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_4D(), ICLKernel::lws_hint(), arm_compute::test::validation::reference::slice(), Window::slide_window_slice_4D(), and IKernel::window().

◆ validate()

Status validate ( const ITensorInfo input,
const ITensorInfo output,
const Coordinates starts,
const Coordinates ends,
const BiStrides strides,
int32_t  begin_mask,
int32_t  end_mask,
int32_t  shrink_axis_mask 
)
static

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

Note
Supported tensor rank: up to 4
Parameters
[in]inputSource tensor. Data type supported: U8/S8/QASYMM8/U16/S16/QSYMM16/U32/S32/F16/F32
[in]outputDestination tensor. Data type supported: Same as input
[in]startsThe starts of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]endsThe ends of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]stridesThe strides of the dimensions of the input tensor to be sliced. The length must be of rank(input).
[in]begin_maskIf the ith bit of begin_mask is set, starts[i] is ignored and the fullest possible range in that dimension is used instead.
[in]end_maskIf the ith bit of end_mask is set, ends[i] is ignored and the fullest possible range in that dimension is used instead.
[in]shrink_axis_maskIf the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. A slice of size 1 starting from starts[i] in the dimension must be preserved.

Definition at line 182 of file CLStridedSliceKernel.cpp.

185 {
186  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(input, output, starts, ends, strides, begin_mask, end_mask, shrink_axis_mask));
187  ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(input->clone().get(), output->clone().get(),
188  starts, ends, strides, begin_mask, end_mask, shrink_axis_mask)
189  .first);
190 
191  return Status{};
192 }
std::pair< Status, Window > validate_and_configure_window(ITensorInfo *input, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier, const Size2D &dilation)
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
Definition: Error.h:193

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

Referenced by CLSlice::validate(), and CLStridedSlice::validate().


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