24.02.1
|
Go to the documentation of this file.
79 const int vec_size_x = 16 /
src->element_size();
80 const int input_width_x =
src->tensor_shape().x();
81 const bool multi_access_x = (input_width_x / vec_size_x > 0);
121 scale_to_apply /= qinfo_in.
scale;
125 offset_to_apply -=
static_cast<int32_t
>(
static_cast<float>(qinfo_in.
offset) * qinfo_in.
scale /
qinfo.
scale);
138 std::pair<int, int> min_max_quant_values =
152 ICLKernel::configure_internal(win);
176 unsigned int idx = 0;
Class to describe a number of elements in each dimension.
std::string to_string(T &&value)
Convert integer and float values to string.
@ QASYMM16
quantized, asymmetric fixed-point 16-bit number
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)
@ QASYMM8
quantized, asymmetric fixed-point 8-bit number unsigned
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(...)
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
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.
ITensor * get_tensor(int id)
Get tensor of a given id from the pac.
void configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst)
Set the input, output.
#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context.
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
const ITensor * get_const_tensor(int id) const
Get constant tensor of a given id.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
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.
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.
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.
#define ARM_COMPUTE_RETURN_ERROR_ON_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.
const std::vector< float > & scale() const
Scale vector accessor.
@ QASYMM8_SIGNED
quantized, asymmetric fixed-point 8-bit number signed
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)
std::string float_to_string_with_full_precision(float val)
Create a string with the float in full precision.
bool slide_window_slice_3D(Window &slice) const
Slide the passed 3D window slice.
static Status validate(const ITensorInfo *src, const ITensorInfo *dst)
Static function to check if given info will lead to a valid configuration.
Describe one of the image's dimensions with a start, end and step.
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.
Window first_slice_window_3D() const
First 3D slice of the window.
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.
Describe a multidimensional execution window.
@ ELEMENTWISE
Elementwise CL kernel type.
Copyright (c) 2017-2024 Arm Limited.
@ F16
16-bit floating-point number
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.
bool is_data_type_float(DataType dt)
Check if a given data type is of floating point type.
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
bool is_data_type_quantized_asymmetric(DataType dt)
Check if a given data type is of asymmetric quantized type.
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Store the tensor's metadata.
@ F32
32-bit floating-point number
constexpr int end() const
Return the end of the dimension.
constexpr const Dimension & x() const
Alias to access the first dimension of the window.
DataType
Available data types.
std::unordered_map< const ITensorInfo *, PaddingSize > get_padding_info(std::initializer_list< const ITensorInfo * > infos)
Stores padding information before configuring a kernel.
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)
const QuantizationInfo qinfo
const std::vector< int32_t > & offset() const
Offset vector accessor.
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.