41 Status validate_arguments(
const ITensorInfo *
input,
const ITensorInfo *output,
float epsilon)
50 if((output !=
nullptr) && (output->total_size() != 0))
60 : _input(nullptr), _output(nullptr), _run_in_place(false)
74 _run_in_place = (output ==
nullptr) || (output == input);
94 build_opts.add_option_if(_run_in_place,
"-DIN_PLACE");
97 _kernel =
create_kernel(compile_context,
"mean_stddev_normalization", build_opts.options());
101 ICLKernel::configure_internal(win);
104 _config_id =
"mean_stddev_normalization_layer_";
129 unsigned int idx = 0;
Window first_slice_window_2D() const
First 2D slice of the window.
static Status validate(const ITensorInfo *input, const ITensorInfo *output=nullptr, float epsilon=1e-8f)
Static function to check if given info will lead to a valid configuration of CLMeanStdDevNormalizatio...
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
#define ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(tensor)
const Window & window() const
The maximum window the kernel can be executed on.
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 ...
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
CLMeanStdDevNormalizationKernel()
Default constructor.
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.
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
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
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
Store the tensor's metadata.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context...
std::string lower_string(const std::string &val)
Lower a given string.
bool slide_window_slice_2D(Window &slice) const
Slide the passed 2D window slice.
Copyright (c) 2017-2021 Arm Limited.
1 channel, 1 F16 per channel
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
void add_option(std::string option)
Adds option to the existing build option list.
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::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
std::string float_to_string_with_full_precision(float val)
Create a string with the float in full precision.
Class to describe a number of elements in each dimension.
unsigned int num_elems_processed_per_iteration
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
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.
virtual size_t element_size() const =0
Element size in bytes calculated as data_size() * num_channels()
Elementeise CL kernel type.
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
void set_dimension_step(size_t dimension, int step)
Set the step of a given dimension.
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...
Interface for OpenCL tensor.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(...)
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)
#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Wrapper to configure the Khronos OpenCL C++ header.
#define ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, msg)
If the condition is true, an error is returned.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
unsigned int adjust_vec_size(unsigned int vec_size, size_t dim0)
Returns the adjusted vector size in case it is less than the input's first dimension, getting rounded down to its closest valid vector size.
void run(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.
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)
void configure(ICLTensor *input, ICLTensor *output=nullptr, float epsilon=1e-8f)
Initialise the kernel's input and outputs.