24.02.1
|
Go to the documentation of this file.
56 float extrapolation_value,
63 _batch_index = batch_index;
64 _extrapolation_value = extrapolation_value;
66 const uint32_t vec_size_x = 4;
70 if (dst_window !=
nullptr)
77 const bool multi_access_x = dst_width_x >= vec_size_x;
78 const bool remainder_x = dst_width_x % vec_size_x > 0;
85 ICLKernel::configure_internal(win);
92 "-DLAST_ACCESSED_X=" +
103 uint32_t batch_index,
104 float extrapolation_value,
114 start.
x >=
static_cast<int32_t
>(
src->dimension(1)) || start.
y >=
static_cast<int32_t
>(
src->dimension(2)) ||
115 end.x >=
static_cast<int32_t
>(
src->dimension(1)) ||
end.y >=
static_cast<int32_t
>(
src->dimension(2)));
117 if (dst_window !=
nullptr)
121 if (
dst->total_size() > 0)
146 unsigned int idx = 0;
std::string to_string(T &&value)
Convert integer and float values to string.
constexpr int start() const
Return the start of the dimension.
SimpleTensor< float > src
const StringSet & options() const
Gets the current options list set.
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
@ NHWC
Num samples, height, width, channels.
constexpr int step() const
Return the step of the dimension.
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
void use_tensor_dimensions(const TensorShape &shape, size_t first_dimension=Window::DimX)
Use the tensor's dimensions to fill the window dimensions.
constexpr size_t num_iterations(size_t dimension) const
Return the number of iterations needed to iterate through a given dimension.
ITensor * get_tensor(int id)
Get tensor of a given id from the pac.
Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
const ITensor * get_const_tensor(int id) const
Get constant tensor of a given id.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_NOT_IN(t,...)
void add_option(std::string option)
Adds option to the existing build option list.
#define ARM_COMPUTE_RETURN_ERROR_ON(cond)
If the condition is true, an error is returned.
#define ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(tensor)
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.
void add_option_if(bool cond, std::string option)
Adds option if a given condition is true;.
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
static Status validate(const ITensorInfo *src, const ITensorInfo *dst, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value=0, Window *dst_window=nullptr)
Static function to check if given info will lead to a valid configuration.
void add_argument(unsigned int &idx, T value)
Add the passed parameters to the object's kernel's arguments starting from the index idx.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT(...)
void configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value=0, Window *dst_window=nullptr)
Configure kernel.
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Describe one of the image's dimensions with a start, end and step.
#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_LAYOUT_NOT_IN(t,...)
void set(size_t dimension, const Dimension &dim)
Set the values of a given dimension.
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.
const Window & window() const
The maximum window the kernel can be executed on.
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
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.
void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override
Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue.
Describe a multidimensional execution window.
@ ELEMENTWISE
Elementwise CL kernel type.
Copyright (c) 2017-2024 Arm Limited.
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
void end(TokenStream &in, bool &valid)
Store the tensor's metadata.
@ F32
32-bit floating-point number
@ UNKNOWN
Unknown data type.
constexpr int end() const
Return the end of the dimension.
constexpr const Dimension & x() const
Alias to access the first dimension of the window.
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.