Compute Library
 23.08
batch_to_space.cl
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018-2021, 2023 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  */
24 #include "helpers.h"
25 
26 #if defined(DATA_TYPE) && defined(BATCH_SIZE)
27 /** Batch to space transformation. (NCHW)
28  *
29  * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
30  * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
31  * @note The input tensor batch size must be passed at compile time using -DBATCH_SIZE. e.g. -DBATCH_SIZE=2
32  *
33  * @deprecated This method for dynamic block shape is not fully mature and will be removed in 23.08 release
34  *
35  * @param[in] input_ptr Pointer to the source tensor. Supported data types: All
36  * @param[in] input_stride_x Stride of the source tensor in X dimension (in bytes)
37  * @param[in] input_step_x input_stride_x * number of elements along X processed per workitem(in bytes)
38  * @param[in] input_stride_y Stride of the source tensor in Y dimension (in bytes)
39  * @param[in] input_step_y input_stride_y * number of elements along Y processed per workitem(in bytes)
40  * @param[in] input_stride_z Stride of the source tensor in Z dimension (in bytes)
41  * @param[in] input_step_z input_stride_z * number of elements along Z processed per workitem(in bytes)
42  * @param[in] input_offset_first_element_in_bytes The offset of the first element in the first source tensor
43  * @param[in] batch_id The input tensor batch id
44  * @param[in] block_shape_ptr Pointer to the source tensor. Supported data types: S32
45  * @param[in] block_shape_stride_x Stride of the source tensor in X dimension (in bytes)
46  * @param[in] block_shape_step_x block_shape_stride_x * number of elements along X processed per workitem(in bytes)
47  * @param[in] block_shape_stride_y Stride of the source tensor in Y dimension (in bytes)
48  * @param[in] block_shape_step_y block_shape_stride_y * number of elements along Y processed per workitem(in bytes)
49  * @param[in] input_offset_first_element_in_bytes The offset of the first element in the first source tensor
50  * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr
51  * @param[in] output_stride_x Stride of the destination tensor in X dimension (in bytes)
52  * @param[in] output_step_x output_stride_x * number of elements along X processed per workitem(in bytes)
53  * @param[in] output_stride_y Stride of the destination tensor in Y dimension (in bytes)
54  * @param[in] output_step_y output_stride_y * number of elements along Y processed per workitem(in bytes)
55  * @param[in] output_stride_z Stride of the source tensor in Z dimension (in bytes)
56  * @param[in] output_step_z output_stride_z * number of elements along Z processed per workitem(in bytes)
57  * @param[in] output_offset_first_element_in_bytes The offset of the first element in the destination tensor
58  */
59 __kernel void batch_to_space_nchw(
60  TENSOR4D_DECLARATION(input),
61  const int batch_id,
62  VECTOR_DECLARATION(block_shape),
63  TENSOR3D_DECLARATION(output))
64 {
65  Tensor4D in = CONVERT_TO_TENSOR4D_STRUCT_NO_STEP(input, 0);
66  Tensor3D out = CONVERT_TO_TENSOR3D_STRUCT(output);
67  Vector block = CONVERT_TO_VECTOR_STRUCT_NO_STEP(block_shape);
68 
69  const int block_x = *((__global int *)vector_offset(&block, 0));
70  const int block_y = *((__global int *)vector_offset(&block, 1));
71 
72  const int x = get_global_id(0);
73  const int y = get_global_id(1);
74  const int z = get_global_id(2);
75 
76  const int in_batch = batch_id + ((x % block_x) + (y % block_y) * block_x) * BATCH_SIZE;
77  const int in_x = x / block_x;
78  const int in_y = y / block_y;
79 
80  *((__global DATA_TYPE *)out.ptr) = *((__global DATA_TYPE *)tensor4D_offset(&in, in_x, in_y, z, in_batch));
81 }
82 #endif // defined(DATA_TYPE) && defined(BATCH_SIZE)
83 
84 #if defined(DATA_TYPE) && defined(BATCH_SIZE) && defined(BLOCK_SHAPE_X) && defined(BLOCK_SHAPE_Y)
85 /** Batch to space transformation. (NCHW)
86  *
87  * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
88  * @note The input tensor batch size must be passed at compile time using -DBATCH_SIZE. e.g. -DBATCH_SIZE=2
89  * @note The block shape x must be passed at compile time using -DBLOCK_SHAPE_X. e.g. -DBLOCK_SHAPE_X=2
90  * @note The block shape y must be passed at compile time using -DBLOCK_SHAPE_Y. e.g. -DBLOCK_SHAPE_Y=2
91  *
92  * @param[in] input_ptr Pointer to the source tensor. Supported data types: All
93  * @param[in] input_stride_x Stride of the source tensor in X dimension (in bytes)
94  * @param[in] input_step_x input_stride_x * number of elements along X processed per workitem(in bytes)
95  * @param[in] input_stride_y Stride of the source tensor in Y dimension (in bytes)
96  * @param[in] input_step_y input_stride_y * number of elements along Y processed per workitem(in bytes)
97  * @param[in] input_stride_z Stride of the source tensor in Z dimension (in bytes)
98  * @param[in] input_step_z input_stride_z * number of elements along Z processed per workitem(in bytes)
99  * @param[in] input_offset_first_element_in_bytes The offset of the first element in the first source tensor
100  * @param[in] batch_id The input tensor batch id
101  * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr
102  * @param[in] output_stride_x Stride of the destination tensor in X dimension (in bytes)
103  * @param[in] output_step_x output_stride_x * number of elements along X processed per workitem(in bytes)
104  * @param[in] output_stride_y Stride of the destination tensor in Y dimension (in bytes)
105  * @param[in] output_step_y output_stride_y * number of elements along Y processed per workitem(in bytes)
106  * @param[in] output_stride_z Stride of the source tensor in Z dimension (in bytes)
107  * @param[in] output_step_z output_stride_z * number of elements along Z processed per workitem(in bytes)
108  * @param[in] output_offset_first_element_in_bytes The offset of the first element in the destination tensor
109  */
110 __kernel void batch_to_space_static_nchw(
111  TENSOR4D_DECLARATION(input),
112  const int batch_id,
113  TENSOR3D_DECLARATION(output))
114 {
115  Tensor4D in = CONVERT_TO_TENSOR4D_STRUCT_NO_STEP(input, 0);
116  Tensor3D out = CONVERT_TO_TENSOR3D_STRUCT(output);
117 
118  const int block_x = BLOCK_SHAPE_X;
119  const int block_y = BLOCK_SHAPE_Y;
120 
121  const int x = get_global_id(0) + CROP_LEFT;
122  const int y = get_global_id(1) + CROP_TOP;
123  const int z = get_global_id(2);
124 
125  const int in_batch = batch_id + ((x % block_x) + (y % block_y) * block_x) * BATCH_SIZE;
126  const int in_x = x / block_x;
127  const int in_y = y / block_y;
128 
129  *((__global DATA_TYPE *)out.ptr) = *((__global DATA_TYPE *)tensor4D_offset(&in, in_x, in_y, z, in_batch));
130 }
131 #endif // defined(DATA_TYPE) && defined(BATCH_SIZE) && defined(BLOCK_SHAPE_X) && defined(BLOCK_SHAPE_Y)
Vector
Structure to hold Vector information.
Definition: helpers.h:917
CONVERT_TO_TENSOR3D_STRUCT
#define CONVERT_TO_TENSOR3D_STRUCT(name)
Definition: helpers.h:898
CONVERT_TO_TENSOR4D_STRUCT_NO_STEP
#define CONVERT_TO_TENSOR4D_STRUCT_NO_STEP(name, mod_size)
Definition: helpers.h:909
Tensor3D
Structure to hold 3D tensor information.
Definition: helpers.h:934
CONVERT_TO_VECTOR_STRUCT_NO_STEP
#define CONVERT_TO_VECTOR_STRUCT_NO_STEP(name)
Definition: helpers.h:880
TENSOR3D_DECLARATION
#define TENSOR3D_DECLARATION(name)
Definition: helpers.h:841
VECTOR_DECLARATION
#define VECTOR_DECLARATION(name)
Definition: helpers.h:827
Tensor4D
Structure to hold 4D tensor information.
Definition: helpers.h:944
tensor4D_offset
const __global uchar * tensor4D_offset(const Tensor4D *tensor, int x, int y, int z, int w)
Get the pointer position of a Tensor4D.
Definition: helpers.h:1137
Tensor3D::ptr
__global uchar * ptr
Pointer to the starting postion of the buffer.
Definition: helpers.h:936
vector_offset
const __global uchar * vector_offset(const Vector *vec, int x)
Get the pointer position of a Vector.
Definition: helpers.h:1101
TENSOR4D_DECLARATION
#define TENSOR4D_DECLARATION(name)
Definition: helpers.h:851
arm_compute::test::validation::input
auto input
Definition: LSTMLayerQuantized.cpp:486
helpers.h