64 std::set<std::string> build_opts;
66 build_opts.insert(
"#define " + dt_name);
70 build_opts.insert(
"#define SOFTMAX_LAYER_MAX");
75 build_opts.insert(
"#define NON_MULTIPLE_OF_8");
88 unsigned int num_elems_written_per_iteration = 1;
91 num_elems_written_per_iteration = 2;
102 IGCKernel::configure(win);
106 : _input(nullptr), _max(nullptr), _output(nullptr), _sum(nullptr)
129 std::set<std::string> build_opts;
131 build_opts.insert(
"#define " + dt_name);
135 build_opts.insert(
"#define SOFTMAX_LAYER_SHIFT_EXP_SUM");
140 build_opts.insert(
"#define NON_MULTIPLE_OF_8");
153 unsigned int num_elems_written_per_iteration = 1;
156 num_elems_written_per_iteration = 2;
171 IGCKernel::configure(win);
186 unsigned int idx = 0;
187 unsigned int binding = 1;
193 _kernel.update_shader_params();
200 : _input(nullptr), _sum(nullptr), _output(nullptr)
220 std::set<std::string> build_opts;
222 build_opts.insert(
"#define " + dt_name);
226 build_opts.insert(
"#define SOFTMAX_LAYER_NORM");
233 unsigned int num_elems_written_per_iteration = 1;
236 num_elems_written_per_iteration = 2;
249 IGCKernel::configure(win);
267 unsigned int idx = 0;
268 unsigned int binding = 1;
274 _kernel.update_shader_params();
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
const Window & window() const
The maximum window the kernel can be executed on.
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...
void enqueue(IGCKernel &kernel, const Window &window, const gles::NDRange &lws=gles::NDRange(1U, 1U, 1U))
Add the kernel to the command queue with the given window.
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
DATA_TYPE sum(__global const DATA_TYPE *input)
Calculate sum of a vector.
std::string to_string(T &&value)
Convert integer and float values to string.
virtual DataType data_type() const =0
Data type used for each element of the tensor.
1 channel, 1 F32 per channel
GCLogits1DShiftExpSumKernel()
Default constructor.
Interface for GLES Compute tensor.
Describe one of the image's dimensions with a start, end and step.
unsigned int num_arguments_per_3D_tensor() const
Returns the number of arguments enqueued per 3D tensor object.
Copyright (c) 2017-2021 Arm Limited.
1 channel, 1 F16 per channel
virtual ValidRegion valid_region() const =0
Valid region of the tensor.
Implementation of a static rectangular access pattern.
void run(const Window &window) override
Enqueue the OpenGL ES shader to process the given window.
void run(const Window &window) override
Enqueue the OpenGL ES shader to process the given window.
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
bool update_window_and_padding(Window &win, Ts &&... patterns)
Update window and padding size for each of the access patterns.
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.
virtual const TensorShape & tensor_shape() const =0
Size for each dimension of the tensor.
#define ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(...)
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.
Manages all the GLES kernels compilation and caching, provides accessors for the GLES Context...
Class to describe a number of elements in each dimension.
Implementation of a row access pattern.
#define ARM_COMPUTE_ERROR_ON_MISMATCHING_SHAPES(...)
bool auto_init_if_empty(ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())
Auto initialize the tensor info (shape, number of channels and data type) if the current assignment i...
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
void set(size_t dimension, const Dimension &dim)
Set the values of a given dimension.
bool slide_window_slice_3D(Window &slice) const
Slide the passed 3D window slice.
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
#define ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Wrapper to configure the Khronos EGL and OpenGL ES C header.
static GCKernelLibrary & get()
Get the static instance of GCKernelLibrary.
#define ARM_COMPUTE_ERROR_ON_MISMATCHING_DIMENSIONS(...)
static constexpr size_t DimZ
Alias for dimension 2 also known as Z dimension.
GCKernel create_kernel(const std::string &shader_name, const StringSet &build_options_set={}) const
Creates a kernel from the kernel library.
unsigned int num_elems_processed_per_iteration
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Container for valid region of a window.
void configure(const IGCTensor *input, IGCTensor *output)
Set the input and output tensors.
void configure(const IGCTensor *input, const IGCTensor *max, IGCTensor *output, IGCTensor *sum)
Set the input and output tensors.
Window first_slice_window_3D() const
First 3D slice of the window.
void configure(const IGCTensor *input, const IGCTensor *sum, IGCTensor *output)
Set the input and output tensors.
Describe a multidimensional execution window.
TensorShape & set(size_t dimension, size_t value, bool apply_dim_correction=true, bool increase_dim_unit=true)
Accessor to set the value of one of the dimensions.
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
GCLogits1DNormKernel()
Default constructor.
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)