1 /* 2 * Copyright (c) 2017 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 #ifndef ARM_COMPUTE_TEST_DERIVATIVE_FIXTURE 25 #define ARM_COMPUTE_TEST_DERIVATIVE_FIXTURE 26 27 #include "tests/Globals.h" 28 #include "tests/IAccessor.h" 29 #include "tests/Types.h" 30 #include "tests/framework/Asserts.h" 31 #include "tests/framework/Fixture.h" 32 #include "tests/validation/reference/Derivative.h" 33 34 #include <memory> 35 36 namespace arm_compute 37 { 38 namespace test 39 { 40 namespace validation 41 { 42 template <typename TensorType, typename AccessorType, typename FunctionType, typename T, typename U> 43 class DerivativeValidationFixture : public framework::Fixture 44 { 45 public: 46 template <typename...> setup(TensorShape shape,BorderMode border_mode,Format format,GradientDimension gradient_dimension)47 void setup(TensorShape shape, BorderMode border_mode, Format format, GradientDimension gradient_dimension) 48 { 49 // Generate a random constant value 50 std::mt19937 gen(library->seed()); 51 std::uniform_int_distribution<uint8_t> int_dist(0, 255); 52 const uint8_t constant_border_value = int_dist(gen); 53 54 _border_mode = border_mode; 55 _target = compute_target(shape, border_mode, format, constant_border_value, gradient_dimension); 56 _reference = compute_reference(shape, border_mode, format, constant_border_value, gradient_dimension); 57 } 58 59 protected: 60 template <typename V> fill(V && tensor)61 void fill(V &&tensor) 62 { 63 library->fill_tensor_uniform(tensor, 0); 64 } 65 66 template <typename V> fill_zero(V && tensor)67 void fill_zero(V &&tensor) 68 { 69 library->fill_tensor_uniform(tensor, 0, static_cast<U>(0), static_cast<U>(0)); 70 } 71 compute_target(const TensorShape & shape,BorderMode border_mode,Format format,uint8_t constant_border_value,GradientDimension gradient_dimension)72 std::pair<TensorType, TensorType> compute_target(const TensorShape &shape, BorderMode border_mode, Format format, uint8_t constant_border_value, GradientDimension gradient_dimension) 73 { 74 // Create tensors 75 TensorType src = create_tensor<TensorType>(shape, data_type_from_format(format)); 76 TensorType dst_x = create_tensor<TensorType>(shape, data_type_from_format(Format::S16)); 77 TensorType dst_y = create_tensor<TensorType>(shape, data_type_from_format(Format::S16)); 78 79 src.info()->set_format(format); 80 dst_x.info()->set_format(Format::S16); 81 dst_y.info()->set_format(Format::S16); 82 83 FunctionType derivative; 84 85 switch(gradient_dimension) 86 { 87 case GradientDimension::GRAD_X: 88 derivative.configure(&src, &dst_x, nullptr, border_mode, constant_border_value); 89 break; 90 case GradientDimension::GRAD_Y: 91 derivative.configure(&src, nullptr, &dst_y, border_mode, constant_border_value); 92 break; 93 case GradientDimension::GRAD_XY: 94 derivative.configure(&src, &dst_x, &dst_y, border_mode, constant_border_value); 95 break; 96 default: 97 ARM_COMPUTE_ERROR("Gradient dimension not supported"); 98 } 99 100 ARM_COMPUTE_EXPECT(src.info()->is_resizable(), framework::LogLevel::ERRORS); 101 ARM_COMPUTE_EXPECT(dst_x.info()->is_resizable(), framework::LogLevel::ERRORS); 102 ARM_COMPUTE_EXPECT(dst_y.info()->is_resizable(), framework::LogLevel::ERRORS); 103 104 // Allocate tensors 105 src.allocator()->allocate(); 106 dst_x.allocator()->allocate(); 107 dst_y.allocator()->allocate(); 108 109 ARM_COMPUTE_EXPECT(!src.info()->is_resizable(), framework::LogLevel::ERRORS); 110 ARM_COMPUTE_EXPECT(!dst_x.info()->is_resizable(), framework::LogLevel::ERRORS); 111 ARM_COMPUTE_EXPECT(!dst_y.info()->is_resizable(), framework::LogLevel::ERRORS); 112 113 // Fill tensors 114 fill(AccessorType(src)); 115 fill_zero(AccessorType(dst_x)); 116 fill_zero(AccessorType(dst_y)); 117 118 // Compute function 119 derivative.run(); 120 121 return std::make_pair(std::move(dst_x), std::move(dst_y)); 122 } 123 compute_reference(const TensorShape & shape,BorderMode border_mode,Format format,uint8_t constant_border_value,GradientDimension gradient_dimension)124 std::pair<SimpleTensor<U>, SimpleTensor<U>> compute_reference(const TensorShape &shape, BorderMode border_mode, Format format, uint8_t constant_border_value, GradientDimension gradient_dimension) 125 { 126 // Create reference 127 SimpleTensor<T> src{ shape, format }; 128 129 // Fill reference 130 fill(src); 131 132 return reference::derivative<U>(src, border_mode, constant_border_value, gradient_dimension); 133 } 134 135 BorderMode _border_mode{ BorderMode::UNDEFINED }; 136 std::pair<TensorType, TensorType> _target{}; 137 std::pair<SimpleTensor<U>, SimpleTensor<U>> _reference{}; 138 }; 139 } // namespace validation 140 } // namespace test 141 } // namespace arm_compute 142 #endif /* ARM_COMPUTE_TEST_DERIVATIVE_FIXTURE */ 143