Compute Library
 19.08
CLWeightsReshapeKernel Class Reference

OpenCL kernel to perform reshaping on the weights used by convolution and locally connected layer. More...

#include <CLWeightsReshapeKernel.h>

Collaboration diagram for CLWeightsReshapeKernel:
[legend]

Public Member Functions

 CLWeightsReshapeKernel ()
 Constructor. More...
 
 CLWeightsReshapeKernel (const CLWeightsReshapeKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
CLWeightsReshapeKerneloperator= (const CLWeightsReshapeKernel &)=delete
 Prevent instances of this class from being copied (As this class contains pointers) More...
 
 CLWeightsReshapeKernel (CLWeightsReshapeKernel &&)=default
 Allow instances of this class to be moved. More...
 
CLWeightsReshapeKerneloperator= (CLWeightsReshapeKernel &&)=default
 Allow instances of this class to be moved. More...
 
 ~CLWeightsReshapeKernel ()=default
 Default destructor. More...
 
void configure (const ICLTensor *input, const ICLTensor *biases, ICLTensor *output, unsigned int num_groups=1)
 Set the input and output of the kernel. More...
 
void run (const Window &window, cl::CommandQueue &queue) override
 Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue. More...
 
- Public Member Functions inherited from ICLKernel
 ICLKernel ()
 Constructor. More...
 
cl::Kernel & kernel ()
 Returns a reference to the OpenCL kernel of this object. More...
 
template<typename T >
void add_1D_array_argument (unsigned int &idx, const ICLArray< T > *array, const Strides &strides, unsigned int num_dimensions, const Window &window)
 Add the passed 1D array's parameters to the object's kernel's arguments starting from the index idx. More...
 
void add_1D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, 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_1D_tensor_argument_if (bool cond, unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 1D tensor's parameters to the object's kernel's arguments starting from the index idx if the condition is true. More...
 
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. More...
 
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 if the condition is true. More...
 
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. More...
 
void add_4D_tensor_argument (unsigned int &idx, const ICLTensor *tensor, const Window &window)
 Add the passed 4D tensor's parameters to the object's kernel's arguments starting from the index idx. More...
 
template<typename T >
void add_argument (unsigned int &idx, T value)
 Add the passed parameters to the object's kernel's arguments starting from the index idx. More...
 
void set_lws_hint (const cl::NDRange &lws_hint)
 Set the Local-Workgroup-Size hint. More...
 
cl::NDRange lws_hint () const
 Return the Local-Workgroup-Size hint. More...
 
const std::string & config_id () const
 Get the configuration ID. More...
 
void set_target (GPUTarget target)
 Set the targeted GPU architecture. More...
 
void set_target (cl::Device &device)
 Set the targeted GPU architecture according to the CL device. More...
 
GPUTarget get_target () const
 Get the targeted GPU architecture. More...
 
size_t get_max_workgroup_size ()
 Get the maximum workgroup size for the device the CLKernelLibrary uses. More...
 
template<typename T , unsigned int dimension_size>
void add_array_argument (unsigned &idx, const ICLArray< T > *array, const Strides &strides, unsigned int num_dimensions, const Window &window)
 Add the passed array's parameters to the object's kernel's arguments starting from the index idx. More...
 
template<unsigned int dimension_size>
void add_tensor_argument (unsigned &idx, const ICLTensor *tensor, const Window &window)
 
- 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...
 
virtual BorderSize border_size () const
 The size of the border for that kernel. More...
 
const Windowwindow () const
 The maximum window the kernel can be executed on. More...
 

Static Public Member Functions

static Status validate (const ITensorInfo *input, const ITensorInfo *biases, const ITensorInfo *output, unsigned int num_groups=1)
 Static function to check if given info will lead to a valid configuration of CLWeightsReshapeKernel. More...
 
- Static Public Member Functions inherited from ICLKernel
static constexpr unsigned int num_arguments_per_1D_array ()
 Returns the number of arguments enqueued per 1D array object. More...
 
static constexpr unsigned int num_arguments_per_1D_tensor ()
 Returns the number of arguments enqueued per 1D tensor object. More...
 
static constexpr unsigned int num_arguments_per_2D_tensor ()
 Returns the number of arguments enqueued per 2D tensor object. More...
 
static constexpr unsigned int num_arguments_per_3D_tensor ()
 Returns the number of arguments enqueued per 3D tensor object. More...
 
static constexpr unsigned int num_arguments_per_4D_tensor ()
 Returns the number of arguments enqueued per 4D tensor object. More...
 
static cl::NDRange gws_from_window (const Window &window)
 Get the global work size given an execution window. More...
 

Detailed Description

OpenCL kernel to perform reshaping on the weights used by convolution and locally connected layer.

Rearranges each 3-dimensional kernel to a single row leading to a matrix with linearized kernels. In combination with the CLIm2ColKernel can transform a convolution to a matrix multiplication.

For example assuming a 3D weight kernel of 3x3 dimensions and depth of 2 we have:

\[ \left( \begin{array}{ccc} a000 & a001 & a002 \\ a010 & a011 & a012 \\ a020 & a021 & a022 \\ \end{array} \right) \left( \begin{array}{ccc} a100 & a101 & a102 \\ a110 & a111 & a112 \\ a120 & a121 & a122 \\ \end{array} \right) \rightarrow \left( \begin{array}{ccccccccc} a000 & a001 & a002 & a010 & a011 & a012 & a020 & a021 & a022 & a100 & a101 & a102 & a110 & a111 & a112 & a120 & a121 & a122 \\ \end{array} \right) \]

Definition at line 54 of file CLWeightsReshapeKernel.h.

Constructor & Destructor Documentation

◆ CLWeightsReshapeKernel() [1/3]

Constructor.

Definition at line 73 of file CLWeightsReshapeKernel.cpp.

74  : _input(nullptr), _biases(nullptr), _output(nullptr)
75 {
76 }

◆ CLWeightsReshapeKernel() [2/3]

Prevent instances of this class from being copied (As this class contains pointers)

◆ CLWeightsReshapeKernel() [3/3]

Allow instances of this class to be moved.

◆ ~CLWeightsReshapeKernel()

~CLWeightsReshapeKernel ( )
default

Default destructor.

Member Function Documentation

◆ configure()

void configure ( const ICLTensor input,
const ICLTensor biases,
ICLTensor output,
unsigned int  num_groups = 1 
)

Set the input and output of the kernel.

Parameters
[in]inputThe input tensor to convert. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM] if shared, and 5D tensor with dimensions [kernel_x, kernel_y, IFM, OFM, num_patches] if unshared. Data types supported: QASYMM8/F16/F32
[in]biasesThe shared biases tensor to append. Bias is 1D tensor with dimensions [OFM] if shared and 2D tensor with dimensions [OFM, num_patches] if unshared. Data types supported: Same as input
Warning
Appending biases to weights reshaped matrix is not supported for quantized asymmetric types.
Parameters
[out]outputThe output tensor. Should be a 2D Tensor if there are no groups and the weights are not shared; a 3D Tensor otherwise. Data types supported: Same as input
[in]num_groups(Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout Number of groups greater than one are only supported for NCHW data layout, and the number of weights must be a multiple of it.

Definition at line 78 of file CLWeightsReshapeKernel.cpp.

79 {
80  ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
81 
82  // Output tensor auto inizialitation if not yet initialized
83  auto_init_if_empty(*output->info(), input->info()->clone()->set_tensor_shape(compute_weights_reshaped_shape(*input->info(), (biases != nullptr), num_groups)));
84 
85  // Perform validation step
86  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(),
87  (biases != nullptr) ? biases->info() : nullptr,
88  output->info(), num_groups));
89 
90  const DataType data_type = input->info()->data_type();
91 
92  _biases = biases;
93  _output = output;
94  _input = input;
95 
96  // Create build options
97  CLBuildOptions build_opts;
98  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(data_type));
99  build_opts.add_option("-DNUM_GROUPS=" + support::cpp11::to_string(num_groups));
100  build_opts.add_option_if(biases != nullptr, "-DHAS_BIAS");
101 
102  // Create kernel
103  _kernel = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel("reshape_to_columns", build_opts.options()));
104 
105  // Configure window
106  Window win = calculate_max_window(*input->info(), Steps());
107  // The CLWeightsReshapeKernel doesn't need padding so update_window_and_padding() can be skipped
108  output->info()->set_valid_region(ValidRegion(Coordinates(), output->info()->tensor_shape()));
109  ICLKernel::configure_internal(win);
110 }
const StringSet & options() const
Gets the current options list set.
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.
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:327
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
virtual void set_valid_region(const ValidRegion &valid_region)=0
Set the valid region of the tensor.
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...
Definition: Helpers.inl:201
void add_option(std::string option)
Adds option to the existing build option list.
virtual const TensorShape & tensor_shape() const =0
Size for each dimension of the tensor.
Class to describe a number of elements in each dimension.
Definition: Steps.h:40
const unsigned int num_groups
Definition: Im2Col.cpp:148
Coordinates of an item.
Definition: Coordinates.h:37
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
Definition: CLHelpers.cpp:35
virtual std::unique_ptr< T > clone() const =0
Provide a clone of the current object of class T.
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
void add_option_if(bool cond, std::string option)
Adds option if a given condition is true;.
std::unique_ptr< Kernel > create_kernel()
Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object.
Definition: Helpers.h:86
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161
TensorShape compute_weights_reshaped_shape(const ITensorInfo &weights, bool has_bias=false, unsigned int num_groups=1)
Calculate the reshaped shape of the weights.
Container for valid region of a window.
Definition: Types.h:174
DataType
Available data types.
Definition: Types.h:74
Describe a multidimensional execution window.
Definition: Window.h:39

References CLBuildOptions::add_option(), CLBuildOptions::add_option_if(), ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, arm_compute::auto_init_if_empty(), arm_compute::calculate_max_window(), ICloneable< T >::clone(), arm_compute::misc::shape_calculator::compute_weights_reshaped_shape(), arm_compute::create_kernel(), arm_compute::test::validation::data_type, ITensorInfo::data_type(), CLKernelLibrary::get(), arm_compute::get_cl_type_from_data_type(), ITensor::info(), arm_compute::test::validation::num_groups, CLBuildOptions::options(), ITensorInfo::set_valid_region(), ITensorInfo::tensor_shape(), and arm_compute::support::cpp11::to_string().

Referenced by CLConvolutionLayerReshapeWeights::configure(), and CLLocallyConnectedLayer::configure().

◆ operator=() [1/2]

CLWeightsReshapeKernel& operator= ( const CLWeightsReshapeKernel )
delete

Prevent instances of this class from being copied (As this class contains pointers)

◆ operator=() [2/2]

CLWeightsReshapeKernel& operator= ( CLWeightsReshapeKernel &&  )
default

Allow instances of this class to be moved.

◆ run()

void run ( const Window window,
cl::CommandQueue &  queue 
)
overridevirtual

Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue.

Note
The queue is not flushed by this method, and therefore the kernel will not have been executed by the time this method returns.
Parameters
[in]windowRegion on which to execute the kernel. (Must be a valid region of the window returned by window()).
[in,out]queueCommand queue on which to enqueue the kernel.

Implements ICLKernel.

Definition at line 118 of file CLWeightsReshapeKernel.cpp.

119 {
122 
123  Window out_window;
124  out_window.use_tensor_dimensions(_output->info()->tensor_shape());
125 
126  Window in_slice = window.first_slice_window_3D();
127  Window out_slice = out_window.first_slice_window_2D();
128 
129  Window biases_window;
130  Window biases_slice;
131 
133  idx += (_biases != nullptr) ? num_arguments_per_1D_tensor() : 0;
134  _kernel.setArg<cl_uint>(idx++, _input->info()->dimension(0));
135  _kernel.setArg<cl_uint>(idx++, _input->info()->dimension(1));
136  _kernel.setArg<cl_uint>(idx++, _input->info()->dimension(2));
137  _kernel.setArg<cl_uint>(idx++, _input->info()->dimension(3));
138  _kernel.setArg<cl_uint>(idx++, _output->info()->strides_in_bytes().z());
139 
140  if(_biases != nullptr)
141  {
142  biases_window.use_tensor_dimensions(_biases->info()->tensor_shape());
143  biases_slice = biases_window.first_slice_window_1D();
144  }
145 
146  do
147  {
148  // Set arguments
149  unsigned idx = 0;
150  add_3D_tensor_argument(idx, _input, in_slice);
151  add_2D_tensor_argument(idx, _output, out_slice);
152  if(_biases != nullptr)
153  {
154  add_1D_tensor_argument(idx, _biases, biases_slice);
155  ARM_COMPUTE_UNUSED(biases_window.slide_window_slice_1D(biases_slice));
156  }
157 
158  // Run kernel
159  enqueue(queue, *this, in_slice);
160  }
161  while(window.slide_window_slice_4D(in_slice) && out_window.slide_window_slice_2D(out_slice));
162 }
static constexpr unsigned int num_arguments_per_1D_tensor()
Returns the number of arguments enqueued per 1D tensor object.
Definition: ICLKernel.h:184
Window first_slice_window_2D() const
First 2D slice of the window.
Definition: Window.h:267
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
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
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.
Definition: ICLKernel.h:158
void use_tensor_dimensions(const TensorShape &shape, size_t first_dimension=Window::DimX)
Use the tensor's dimensions to fill the window dimensions.
Definition: Window.inl:250
bool slide_window_slice_2D(Window &slice) const
Slide the passed 2D window slice.
Definition: Window.h:307
static constexpr unsigned int num_arguments_per_3D_tensor()
Returns the number of arguments enqueued per 3D tensor object.
Definition: ICLKernel.h:200
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Definition: Error.h:160
virtual const TensorShape & tensor_shape() const =0
Size for each dimension of the tensor.
T z() const
Alias to access the size of the third dimension.
Definition: Dimensions.h:91
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
#define ARM_COMPUTE_ERROR_ON_MISMATCHING_WINDOWS(f, w)
Definition: Validate.h:183
static constexpr unsigned int num_arguments_per_2D_tensor()
Returns the number of arguments enqueued per 2D tensor object.
Definition: ICLKernel.h:192
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.
Definition: ICLKernel.h:134
bool slide_window_slice_4D(Window &slice) const
Slide the passed 4D window slice.
Definition: Window.h:331
virtual const Strides & strides_in_bytes() const =0
The strides in bytes for accessing each dimension of the tensor.
void add_1D_tensor_argument(unsigned int &idx, const ICLTensor *tensor, const Window &window)
Add the passed 1D tensor's parameters to the object's kernel's arguments starting from the index idx.
Definition: ICLKernel.h:110
Window first_slice_window_3D() const
First 3D slice of the window.
Definition: Window.h:275
bool slide_window_slice_1D(Window &slice) const
Slide the passed 1D window slice.
Definition: Window.h:295
Describe a multidimensional execution window.
Definition: Window.h:39
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:940
Window first_slice_window_1D() const
First 1D slice of the window.
Definition: Window.h:259

References ICLKernel::add_1D_tensor_argument(), ICLKernel::add_2D_tensor_argument(), ICLKernel::add_3D_tensor_argument(), ARM_COMPUTE_ERROR_ON_MISMATCHING_WINDOWS, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, ARM_COMPUTE_UNUSED, ITensorInfo::dimension(), arm_compute::enqueue(), Window::first_slice_window_1D(), Window::first_slice_window_2D(), Window::first_slice_window_3D(), ITensor::info(), ICLKernel::num_arguments_per_1D_tensor(), ICLKernel::num_arguments_per_2D_tensor(), ICLKernel::num_arguments_per_3D_tensor(), Window::slide_window_slice_1D(), Window::slide_window_slice_2D(), Window::slide_window_slice_4D(), ITensorInfo::strides_in_bytes(), ITensorInfo::tensor_shape(), Window::use_tensor_dimensions(), IKernel::window(), and Dimensions< T >::z().

◆ validate()

Status validate ( const ITensorInfo input,
const ITensorInfo biases,
const ITensorInfo output,
unsigned int  num_groups = 1 
)
static

Static function to check if given info will lead to a valid configuration of CLWeightsReshapeKernel.

Parameters
[in]inputThe input tensor to convert. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM] if shared, and 5D tensor with dimensions [kernel_x, kernel_y, IFM, OFM, num_patches] if unshared. Data types supported: QASYMM8/F16/F32
[in]biasesThe shared biases tensor to append. Bias is 1D tensor with dimensions [OFM] if shared and 2D tensor with dimensions [OFM, num_patches] if unshared. Data types supported: Same as input
Warning
Appending biases to weights reshaped matrix is not supported for quantized asymmetric types.
Parameters
[in]outputThe output tensor. Should be a 2D Tensor if there are no groups and the weights are not shared; a 3D Tensor otherwise. Data types supported: Same as input
[in]num_groups(Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout Number of groups greater than one are only supported for NCHW data layout, and the number of weights must be a multiple of it.
Returns
a status

Definition at line 112 of file CLWeightsReshapeKernel.cpp.

113 {
114  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(input, biases, output, num_groups));
115  return Status{};
116 }
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
Definition: Error.h:193
Status class.
Definition: Error.h:52
const unsigned int num_groups
Definition: Im2Col.cpp:148

References ARM_COMPUTE_RETURN_ON_ERROR, and arm_compute::test::validation::num_groups.

Referenced by CLConvolutionLayerReshapeWeights::validate(), and CLLocallyConnectedLayer::validate().


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