Compute Library
 21.02
NEConvertFullyConnectedWeightsKernel.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018-2021 Arm Limited.
3  *
4  * SPDX-License-Identifier: MIT
5  *
6  * Permission is hereby granted, free of charge, to any person obtaining a copy
7  * of this software and associated documentation files (the "Software"), to
8  * deal in the Software without restriction, including without limitation the
9  * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10  * sell copies of the Software, and to permit persons to whom the Software is
11  * furnished to do so, subject to the following conditions:
12  *
13  * The above copyright notice and this permission notice shall be included in all
14  * copies or substantial portions of the Software.
15  *
16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22  * SOFTWARE.
23  */
25 
27 #include "arm_compute/core/Types.h"
30 
31 namespace arm_compute
32 {
34  : _input(nullptr), _output(nullptr), _factor1(0), _factor2(0)
35 {
36 }
37 
38 void NEConvertFullyConnectedWeightsKernel::configure(const ITensor *input, ITensor *output, const TensorShape &original_input_shape,
40 {
41  ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
42 
43  // Output tensor auto initialisation if not yet initialized
44  auto_init_if_empty(*output->info(), *input->info()->clone());
45 
47 
48  _input = input;
49  _output = output;
50 
51  const DataLayout input_data_layout = (data_layout == DataLayout::NCHW) ? DataLayout::NHWC : DataLayout::NCHW;
52 
53  const int width_idx = get_data_layout_dimension_index(input_data_layout, DataLayoutDimension::WIDTH);
54  const int height_idx = get_data_layout_dimension_index(input_data_layout, DataLayoutDimension::HEIGHT);
55  const int channel_idx = get_data_layout_dimension_index(input_data_layout, DataLayoutDimension::CHANNEL);
56 
57  const unsigned int num_elems_per_input_plane = original_input_shape[width_idx] * original_input_shape[height_idx];
58  const unsigned int num_channels = original_input_shape[channel_idx];
59 
60  _factor1 = (data_layout == DataLayout::NCHW) ? num_elems_per_input_plane : num_channels;
61  _factor2 = (data_layout == DataLayout::NCHW) ? num_channels : num_elems_per_input_plane;
62 
63  // Configure kernel window
64  Window win = calculate_max_window(*input->info(), Steps());
65  INEKernel::configure(win);
66 }
67 
70 {
72  //Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use Neon FP16 instructions.
75  ARM_COMPUTE_RETURN_ERROR_ON(input->dimension(1) != original_input_shape.total_size_lower(3));
77 
78  // Checks performed when output is configured
79  if((output != nullptr) && (output->total_size() != 0))
80  {
83  }
84 
85  return Status{};
86 }
87 
88 template <typename T>
89 void NEConvertFullyConnectedWeightsKernel::run_convert_fc_weights(const Window &window)
90 {
91  const unsigned int dst_stride_x = _output->info()->strides_in_bytes().x();
92  const unsigned int dst_stride_y = _output->info()->strides_in_bytes().y();
93 
94  Iterator input(_input, window);
95  Iterator output(_output, window);
96 
97  execute_window_loop(window, [&](const Coordinates & id)
98  {
99  *reinterpret_cast<T *>(output.ptr() + id.x() * dst_stride_x + (id.y() % _factor1 * _factor2 + id.y() / _factor1) * dst_stride_y) = *reinterpret_cast<T *>(input.ptr());
100  },
101  input);
102 }
103 
105 {
106  ARM_COMPUTE_UNUSED(info);
109 
110  switch(_input->info()->element_size())
111  {
112  case 1:
113  run_convert_fc_weights<uint8_t>(window);
114  break;
115  case 2:
116  run_convert_fc_weights<uint16_t>(window);
117  break;
118  case 4:
119  run_convert_fc_weights<uint32_t>(window);
120  break;
121  default:
122  ARM_COMPUTE_ERROR("Data type not supported.");
123  break;
124  }
125 }
126 } // namespace arm_compute
virtual size_t num_dimensions() const =0
The number of dimensions of the tensor (rank)
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.
Definition: IKernel.cpp:28
Shape of a tensor.
Definition: TensorShape.h:39
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
#define ARM_COMPUTE_ERROR(msg)
Print the given message then throw an std::runtime_error.
Definition: Error.h:352
virtual DataType data_type() const =0
Data type used for each element of the tensor.
const DataLayout data_layout
Definition: Im2Col.cpp:151
Store the tensor&#39;s metadata.
Definition: ITensorInfo.h:40
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
Status class.
Definition: Error.h:52
#define ARM_COMPUTE_RETURN_ERROR_ON(cond)
If the condition is true, an error is returned.
Definition: Error.h:296
Interface for Neon tensor.
Definition: ITensor.h:36
size_t total_size_lower(size_t dimension) const
Compute size of dimensions lower than the given one.
Definition: TensorShape.h:194
Copyright (c) 2017-2021 Arm Limited.
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Definition: Validate.h:163
T x() const
Alias to access the size of the first dimension.
Definition: Dimensions.h:87
static Status validate(const ITensorInfo *input, const ITensorInfo *output, const TensorShape &original_input_shape, DataLayout data_layout)
Static function to check if given info will lead to a valid configuration of NEConvertFullyConnectedW...
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Definition: Error.h:152
Class to describe a number of elements in each dimension.
Definition: Steps.h:40
Coordinates of an item.
Definition: Coordinates.h:37
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 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&#39;s metadata.
constexpr uint8_t * ptr() const
Return a pointer to the current pixel.
Definition: Helpers.inl:139
virtual size_t element_size() const =0
Element size in bytes calculated as data_size() * num_channels()
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:941
Num samples, channels, height, width.
ScaleKernelInfo info(interpolation_policy, default_border_mode, PixelValue(), sampling_policy, false)
Information about executing thread and CPU.
Definition: CPPTypes.h:235
virtual size_t total_size() const =0
Returns the total size of the tensor in bytes.
void configure(const ITensor *input, ITensor *output, const TensorShape &original_input_shape, DataLayout data_layout)
Set the input and output tensor.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(...)
Definition: Validate.h:443
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)
Definition: Validate.h:545
Num samples, height, width, channels.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161
void execute_window_loop(const Window &w, L &&lambda_function, Ts &&... iterators)
Iterate through the passed window, automatically adjusting the iterators and calling the lambda_funct...
Definition: Helpers.inl:77
T y() const
Alias to access the size of the second dimension.
Definition: Dimensions.h:92
void run(const Window &window, const ThreadInfo &info) override
Execute the kernel on the passed window.
virtual const Strides & strides_in_bytes() const =0
The strides in bytes for accessing each dimension of the tensor.
size_t get_data_layout_dimension_index(const DataLayout data_layout, const DataLayoutDimension data_layout_dimension)
Get the index of the given dimension.
Definition: Helpers.inl:193
Iterator updated by execute_window_loop for each window element.
Definition: Helpers.h:46
DataLayout
[DataLayout enum definition]
Definition: Types.h:120
Describe a multidimensional execution window.
Definition: Window.h:39
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:205