Compute Library
 19.08
GCScaleKernel Class Reference

Interface for the scale kernel. More...

#include <GCScaleKernel.h>

Collaboration diagram for GCScaleKernel:
[legend]

Public Member Functions

void configure (const IGCTensor *input, IGCTensor *output, InterpolationPolicy policy, bool border_undefined, SamplingPolicy sampling_policy=SamplingPolicy::CENTER)
 Initialise the kernel's inputs, output and interpolation policy. More...
 
void run (const Window &window) override
 Enqueue the OpenGL ES shader to process the given window. More...
 
BorderSize border_size () const override
 The size of the border for that kernel. More...
 
- Public Member Functions inherited from IGCSimpleKernel
 IGCSimpleKernel ()
 Constructor. More...
 
 IGCSimpleKernel (const IGCSimpleKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
IGCSimpleKerneloperator= (const IGCSimpleKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 IGCSimpleKernel (IGCSimpleKernel &&)=default
 Allow instances of this class to be moved. More...
 
IGCSimpleKerneloperator= (IGCSimpleKernel &&)=default
 Allow instances of this class to be moved. More...
 
 ~IGCSimpleKernel ()=default
 Default destructor. More...
 
void configure (const IGCTensor *input, IGCTensor *output, unsigned int num_elems_processed_per_iteration, bool border_undefined=false, const BorderSize &border_size=BorderSize())
 Configure the kernel. More...
 
- Public Member Functions inherited from IGCKernel
 IGCKernel ()
 Constructor. More...
 
GCKernelkernel ()
 Returns a reference to the GLES kernel of this object. More...
 
void add_1D_tensor_argument (unsigned int &idx, const IGCTensor *tensor, const unsigned int binding_point, 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_2D_tensor_argument (unsigned int &idx, const IGCTensor *tensor, const unsigned int binding_point, 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_3D_tensor_argument (unsigned int &idx, const IGCTensor *tensor, const unsigned int binding_point, const Window &window)
 Add the passed 3D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
unsigned int num_arguments_per_1D_tensor () const
 Returns the number of arguments enqueued per 1D tensor object. More...
 
unsigned int num_arguments_per_2D_tensor () const
 Returns the number of arguments enqueued per 2D tensor object. More...
 
unsigned int num_arguments_per_3D_tensor () const
 Returns the number of arguments enqueued per 3D tensor object. More...
 
void set_lws_hint (gles::NDRange &lws_hint)
 Set the Local-Workgroup-Size hint. More...
 
void set_target (GPUTarget target)
 Set the targeted GPU architecture. More...
 
GPUTarget get_target () const
 Get the targeted GPU architecture. 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...
 

Detailed Description

Interface for the scale kernel.

Definition at line 35 of file GCScaleKernel.h.

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 42 of file GCScaleKernel.cpp.

43 {
44  return BorderSize(1);
45 }
Container for 2D border size.
Definition: Types.h:259

Referenced by GCScaleKernel::configure().

◆ configure()

void configure ( const IGCTensor input,
IGCTensor output,
InterpolationPolicy  policy,
bool  border_undefined,
SamplingPolicy  sampling_policy = SamplingPolicy::CENTER 
)

Initialise the kernel's inputs, output and interpolation policy.

Parameters
[in]inputSource tensor. Data types supported: F16
[out]outputDestination tensor. Data types supported: Same as input All but the lowest two dimensions must be the same size as in the input tensor, i.e. scaling is only performed within the XY-plane.
[in]policyInterpolation type to use
[in]border_undefinedTrue if the border mode is undefined. False if it's replicate or constant.
[in]sampling_policy(Optional) Sampling policy used by the interpolation. Defaults to SamplingPolicy::CENTER

Definition at line 47 of file GCScaleKernel.cpp.

48 {
52  ARM_COMPUTE_ERROR_ON(output == input);
54 
55  _input = input;
56  _output = output;
57 
58  // Compute the ratio between source width/height and destination width/height
59  const auto wr = static_cast<float>(input->info()->dimension(0)) / static_cast<float>(output->info()->dimension(0));
60  const auto hr = static_cast<float>(input->info()->dimension(1)) / static_cast<float>(output->info()->dimension(1));
61 
62  // Compute actual border size
63  BorderSize border = border_undefined ? BorderSize(0) : border_size();
64 
65  // Area interpolation behaves as Nearest Neighbour in case of up-sampling
66  if(policy == InterpolationPolicy::AREA && wr <= 1.f && hr <= 1.f)
67  {
69  }
70  else
71  {
73  }
74 
75  // Create kernel
76  std::set<std::string> build_opts;
77  build_opts.emplace("#define LOCAL_SIZE_X " + support::cpp11::to_string(1));
78  build_opts.emplace("#define LOCAL_SIZE_Y " + support::cpp11::to_string(1));
79  build_opts.emplace("#define LOCAL_SIZE_Z " + support::cpp11::to_string(1));
80 
81  build_opts.emplace("#define DATA_TYPE_FP16");
82  build_opts.emplace("#define BORDER_SIZE " + support::cpp11::to_string(border.right));
83  if(sampling_policy == SamplingPolicy::TOP_LEFT)
84  {
85  build_opts.emplace("#define SAMPLING_POLICY_TOP_LEFT");
86  }
87  else
88  {
89  build_opts.emplace("#define SAMPLING_POLICY_CENTER");
90  }
91 
92  // Configure kernel window
93  unsigned int num_elems_processed_per_iteration = 4;
94  unsigned int input_width_alignment = 2;
95 
96  // performance optimization for 2x upscaling with no border
97  if((fabs(wr - 0.5) < 1e-6) && (fabs(hr - 0.5) < 1e-6) && border_undefined)
98  {
99  num_elems_processed_per_iteration = 8;
100  input_width_alignment = 4;
101  build_opts.emplace("#define SCALE_NEAREST_8X");
102  }
103  else
104  {
105  build_opts.emplace("#define SCALE_NEAREST_GENERIC");
106  }
107 
108  std::string interpolation_name = string_from_interpolation_policy(policy); // NOLINT
109  std::transform(interpolation_name.begin(), interpolation_name.end(), interpolation_name.begin(), ::tolower);
110  std::string kernel_name = "scale_" + interpolation_name;
111  _kernel = GCKernelLibrary::get().create_kernel(kernel_name, build_opts);
112 
113  Window win = calculate_max_window(*output->info(), Steps(num_elems_processed_per_iteration));
114 
115  const ValidRegion &input_valid_region = input->info()->valid_region();
116 
117  const int total_width = border.left + input_valid_region.anchor[0] + input_valid_region.shape[0] + border.right;
118  const int padding_right = ceil_to_multiple(total_width, input_width_alignment) - border.left - input_valid_region.anchor[0] - input_valid_region.shape[0];
119 
120  // Reads can occur within the valid region of the input
121  AccessWindowStatic input_access(input->info(),
122  input_valid_region.anchor[0] - border.left, input_valid_region.anchor[1] - border.top,
123  input_valid_region.anchor[0] + input_valid_region.shape[0] + padding_right,
124  input_valid_region.anchor[1] + input_valid_region.shape[1] + border.bottom);
125 
126  AccessWindowHorizontal output_access(output->info(), 0, num_elems_processed_per_iteration);
127 
128  update_window_and_padding(win, input_access, output_access);
129 
130  output_access.set_valid_region(win, calculate_valid_region_scale(*(input->info()),
131  output->info()->tensor_shape(),
132  policy,
133  sampling_policy,
134  border_undefined));
135 
136  IGCKernel::configure(win);
137 
138  unsigned int idx = 2 * num_arguments_per_3D_tensor(); //Skip the tensor parameters
139  _kernel.set_argument<float>(idx++, static_cast<float>(input->info()->dimension(0)));
140  _kernel.set_argument<float>(idx++, static_cast<float>(input->info()->dimension(1)));
141  _kernel.set_argument<float>(idx++, wr);
142  _kernel.set_argument<float>(idx++, hr);
143 }
BorderSize border_size() const override
The size of the border for that kernel.
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
#define ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(...)
Definition: Validate.h:543
Container for 2D border size.
Definition: Types.h:259
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:337
unsigned int num_arguments_per_3D_tensor() const
Returns the number of arguments enqueued per 3D tensor object.
Definition: IGCKernel.cpp:147
Output values are defined to match the source pixel whose center is nearest to the sample position.
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps=Steps(), bool skip_border=false, BorderSize border_size=BorderSize())
Calculate the maximum window for a given tensor shape and border setting.
Definition: Helpers.cpp:28
1 channel, 1 F16 per channel
virtual ValidRegion valid_region() const =0
Valid region of the tensor.
Implementation of a static rectangular access pattern.
std::string tolower(std::string string)
Convert string to lower case.
Definition: Utility.h:188
bool update_window_and_padding(Window &win, Ts &&... patterns)
Update window and padding size for each of the access patterns.
Definition: Helpers.h:402
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
Class to describe a number of elements in each dimension.
Definition: Steps.h:40
Implementation of a row access pattern.
const std::string & string_from_interpolation_policy(InterpolationPolicy policy)
Translates a given interpolation policy to a string.
Definition: Utils.cpp:216
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
Samples are taken at pixel top left corner.
Output values are determined by averaging the source pixels whose areas fall under the area of the de...
#define ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Definition: Validate.h:789
ValidRegion calculate_valid_region_scale(const ITensorInfo &src_info, const TensorShape &dst_shape, InterpolationPolicy interpolate_policy, SamplingPolicy sampling_policy, bool border_undefined)
Helper function to calculate the Valid Region for Scale.
Definition: Helpers.cpp:184
static GCKernelLibrary & get()
Get the static instance of GCKernelLibrary.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161
GCKernel create_kernel(const std::string &shader_name, const StringSet &build_options_set={}) const
Creates a kernel from the kernel library.
Container for valid region of a window.
Definition: Types.h:174
Describe a multidimensional execution window.
Definition: Window.h:39
Coordinates anchor
Anchor for the start of the valid region.
Definition: Types.h:246

References ValidRegion::anchor, arm_compute::AREA, ARM_COMPUTE_ERROR_ON, ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN, ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES, ARM_COMPUTE_ERROR_ON_NULLPTR, GCScaleKernel::border_size(), arm_compute::calculate_max_window(), arm_compute::calculate_valid_region_scale(), arm_compute::ceil_to_multiple(), GCKernelLibrary::create_kernel(), ITensorInfo::dimension(), arm_compute::F16, GCKernelLibrary::get(), ITensor::info(), arm_compute::NEAREST_NEIGHBOR, IGCKernel::num_arguments_per_3D_tensor(), arm_compute::test::validation::policy, arm_compute::string_from_interpolation_policy(), ITensorInfo::tensor_shape(), arm_compute::support::cpp11::to_string(), arm_compute::utility::tolower(), arm_compute::TOP_LEFT, arm_compute::update_window_and_padding(), and ITensorInfo::valid_region().

◆ run()

void run ( const Window window)
overridevirtual

Enqueue the OpenGL ES shader to process the given window.

Parameters
[in]windowRegion on which to execute the kernel. (Must be a valid region of the window returned by window()).

Reimplemented from IGCSimple3DKernel.

Definition at line 145 of file GCScaleKernel.cpp.

146 {
149 
150  _kernel.use();
151 
152  _output->set_needs_shifting(true);
153 
155  Window slice_in = window.first_slice_window_3D();
156 
157  slice.shift(Window::DimX, -(_output->info()->padding()).left);
158 
159  do
160  {
161  unsigned int idx = 0;
162  add_3D_tensor_argument(idx, _input, 1, slice_in);
163  add_3D_tensor_argument(idx, _output, 2, slice);
164  _kernel.update_shader_params();
165  enqueue(*this, slice);
166  }
168 }
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
void add_3D_tensor_argument(unsigned int &idx, const IGCTensor *tensor, const unsigned int binding_point, const Window &window)
Add the passed 3D tensor's parameters to the object's kernel's arguments starting from the index idx.
Definition: IGCKernel.cpp:132
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
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Definition: Window.h:43
bool slide_window_slice_3D(Window &slice) const
Slide the passed 3D window slice.
Definition: Window.h:319
Window first_slice_window_3D() const
First 3D slice of the window.
Definition: Window.h:275
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:205
Describe a multidimensional execution window.
Definition: Window.h:39
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:940
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)

References IGCKernel::add_3D_tensor_argument(), ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, Window::DimX, arm_compute::enqueue(), Window::first_slice_window_3D(), arm_compute::test::validation::reference::slice(), Window::slide_window_slice_3D(), and IKernel::window().


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