• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2  * Copyright (c) 2017-2018 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 "arm_compute/core/WindowIterator.h"
25 #include "tests/Utils.h"
26 #include "tests/framework/Asserts.h"
27 #include "tests/framework/Macros.h"
28 #include "tests/framework/datasets/Datasets.h"
29 #include "tests/validation/Validation.h"
30 #include "utils/TypePrinter.h"
31 
32 #include <stdexcept>
33 
34 using namespace arm_compute;
35 using namespace arm_compute::test;
36 using namespace arm_compute::test::validation;
37 
38 TEST_SUITE(UNIT)
TEST_SUITE(WindowIterator)39 TEST_SUITE(WindowIterator)
40 
41 template <typename Dim, typename... Dims>
42 Window create_window(Dim &&dim0, Dims &&... dims)
43 {
44     Window win;
45     const std::array < Dim, 1 + sizeof...(Dims) > dimensions{ { dim0, std::forward<Dims>(dims)... } };
46     for(size_t i = 0; i < dimensions.size(); i++)
47     {
48         win.set(i, dimensions[i]);
49     }
50     return win;
51 }
52 
53 template <typename T>
create_vector(std::initializer_list<T> list_objs)54 std::vector<T> create_vector(std::initializer_list<T> list_objs)
55 {
56     std::vector<T> vec_objs;
57     for(auto it : list_objs)
58     {
59         vec_objs.push_back(it);
60     }
61     return vec_objs;
62 }
63 
64 DATA_TEST_CASE(WholeWindow, framework::DatasetMode::ALL, zip(framework::dataset::make("Window", { create_window(Window::Dimension(0, 1)),
65                                                                                                   create_window(Window::Dimension(1, 5, 2), Window::Dimension(3, 5)),
66                                                                                                   create_window(Window::Dimension(4, 16, 4), Window::Dimension(3, 13, 5), Window::Dimension(1, 3, 2))
67                                                                                                 }),
68                                                              framework::dataset::make("Expected", { create_vector({ Coordinates(0, 0) }),
69                                                                                                     create_vector({ Coordinates(1, 3), Coordinates(3, 3), Coordinates(1, 4), Coordinates(3, 4) }),
70                                                                                                     create_vector({ Coordinates(4, 3, 1), Coordinates(8, 3, 1), Coordinates(12, 3, 1), Coordinates(4, 8, 1), Coordinates(8, 8, 1), Coordinates(12, 8, 1) })
71                                                                                                   })),
72                window, expected)
73 {
74     unsigned int i            = 0;
75     int          row_size     = 0;
76     TensorShape  window_shape = window.shape();
77     Coordinates  start_offset = index2coords(window_shape, 0);
78     Coordinates  end_offset   = index2coords(window_shape, window.num_iterations_total() - 1);
79     auto window_iterator      = create_window_iterator(window, start_offset, end_offset, [&](const Coordinates & id)
__anon38ab53150102(const Coordinates & id) 80     {
81         ARM_COMPUTE_EXPECT_EQUAL(row_size, (window[0].end() - window[0].start()), framework::LogLevel::ERRORS);
82         ARM_COMPUTE_ASSERT(i < expected.size());
83         Coordinates expected_coords = expected[i++];
84         //Set number of dimensions to the maximum (To match the number of dimensions used by the id passed to the lambda function)
85         expected_coords.set_num_dimensions(Coordinates::num_max_dimensions);
86         ARM_COMPUTE_EXPECT_EQUAL(id, expected_coords, framework::LogLevel::ERRORS);
87     });
88     window_iterator.iterate_3D([&](int start, int end)
__anon38ab53150202(int start, int end) 89     {
90         ARM_COMPUTE_EXPECT_EQUAL(window[0].start(), start, framework::LogLevel::ERRORS);
91         ARM_COMPUTE_EXPECT_EQUAL(window[0].end(), end, framework::LogLevel::ERRORS);
92         ARM_COMPUTE_EXPECT(end > start, framework::LogLevel::ERRORS);
93         row_size = end - start;
94     });
95     ARM_COMPUTE_EXPECT_EQUAL(i, expected.size(), framework::LogLevel::ERRORS);
96 }
97 
98 DATA_TEST_CASE(PartialWindow2D, framework::DatasetMode::ALL, zip(zip(zip(combine(framework::dataset::make("Window",
99                                                                                                           create_window(Window::Dimension(4, 20, 4), Window::Dimension(3, 32, 5), Window::Dimension(1, 2, 1))),
100                                                                                  framework::dataset::make("Start", { 0, 1, 3, 2, 4 })),
101                                                                          framework::dataset::make("End", { 0, 2, 5, 8, 7 })),
102                                                                      framework::dataset::make("RowSize",
103 {
104     create_vector({ 4 }),
105     create_vector({ 8, 8 }),
106     create_vector({ 4, 8, 8 }),
107     create_vector({ 8, 8, 16, 16, 16, 16, 4 }),
108     create_vector({ 16, 16, 16, 16 }),
109 })),
110 framework::dataset::make("Expected", { create_vector({ Coordinates(4, 3, 1) }), create_vector({ Coordinates(8, 3, 1), Coordinates(12, 3, 1) }), create_vector({ Coordinates(16, 3, 1), Coordinates(4, 8, 1), Coordinates(8, 8, 1) }), create_vector({ Coordinates(12, 3, 1), Coordinates(16, 3, 1), Coordinates(4, 8, 1), Coordinates(8, 8, 1), Coordinates(12, 8, 1), Coordinates(16, 8, 1), Coordinates(4, 13, 1) }), create_vector({ Coordinates(4, 8, 1), Coordinates(8, 8, 1), Coordinates(12, 8, 1), Coordinates(16, 8, 1) }) })),
111 window, start, end, expected_row_size, expected)
112 {
113     unsigned int i            = 0;
114     int          row_size     = 0;
115     TensorShape  window_shape = window.shape();
116     Coordinates  start_offset = index2coords(window_shape, start);
117     Coordinates  end_offset   = index2coords(window_shape, end);
118     auto window_iterator      = create_window_iterator(window, start_offset, end_offset, [&](const Coordinates & id)
__anon38ab53150302(const Coordinates & id) 119     {
120         ARM_COMPUTE_ASSERT(i < expected.size());
121         ARM_COMPUTE_EXPECT_EQUAL(expected_row_size[i], row_size, framework::LogLevel::ERRORS);
122         Coordinates expected_coords = expected[i++];
123         //Set number of dimensions to the maximum (To match the number of dimensions used by the id passed to the lambda function)
124         expected_coords.set_num_dimensions(Coordinates::num_max_dimensions);
125         ARM_COMPUTE_EXPECT_EQUAL(id, expected_coords, framework::LogLevel::ERRORS);
126     });
127     window_iterator.iterate_3D([&](int start, int end)
__anon38ab53150402(int start, int end) 128     {
129         ARM_COMPUTE_EXPECT(start >= window[0].start(), framework::LogLevel::ERRORS);
130         ARM_COMPUTE_EXPECT(end <= window[0].end(), framework::LogLevel::ERRORS);
131         ARM_COMPUTE_EXPECT(end > start, framework::LogLevel::ERRORS);
132         row_size = end - start;
133     });
134     ARM_COMPUTE_EXPECT_EQUAL(i, expected.size(), framework::LogLevel::ERRORS);
135 }
136 
137 TEST_SUITE_END()
138 TEST_SUITE_END()
139