diff options
Diffstat (limited to 'src/backends/backendsCommon/test/LayerTests.hpp')
| -rw-r--r-- | src/backends/backendsCommon/test/LayerTests.hpp | 606 |
1 files changed, 565 insertions, 41 deletions
diff --git a/src/backends/backendsCommon/test/LayerTests.hpp b/src/backends/backendsCommon/test/LayerTests.hpp index 19105855e1..8bbd0d47c8 100644 --- a/src/backends/backendsCommon/test/LayerTests.hpp +++ b/src/backends/backendsCommon/test/LayerTests.hpp @@ -10,6 +10,8 @@ #include <Half.hpp> #include "TensorCopyUtils.hpp" #include "WorkloadTestUtils.hpp" +#include "TensorUtils.hpp" +#include "Permute.hpp" #include <backendsCommon/CpuTensorHandle.hpp> #include <backendsCommon/IBackendInternal.hpp> @@ -843,36 +845,41 @@ LayerTestResult<int16_t, 4> TanhInt16Test( const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); -// Tests that the output should be identical to the input when the output dimensions match the input ones. -LayerTestResult<float, 4> ResizeBilinearNopTest( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, - const armnn::DataLayout dataLayout); +/// Tests that the output should be identical to the input when the output dimensions match the input ones. +template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> +LayerTestResult<T, 4> ResizeBilinearNopTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); // Tests the behaviour of the resize bilinear operation when rescaling a 2x2 image into a 1x1 image. -LayerTestResult<float, 4> SimpleResizeBilinearTest( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, - const armnn::DataLayout dataLayout); +template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> +LayerTestResult<T, 4> SimpleResizeBilinearTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); // Tests the resize bilinear for minification of a square input matrix (also: input dimensions are a // multiple of output dimensions). -LayerTestResult<float, 4> ResizeBilinearSqMinTest( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, - const armnn::DataLayout dataLayout); +template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> +LayerTestResult<T, 4> ResizeBilinearSqMinTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); // Tests the resize bilinear for minification (output dimensions smaller than input dimensions). -LayerTestResult<float, 4> ResizeBilinearMinTest( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, - const armnn::DataLayout dataLayout); +template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> +LayerTestResult<T, 4> ResizeBilinearMinTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); // Tests the resize bilinear for magnification (output dimensions bigger than input dimensions). -LayerTestResult<float, 4> ResizeBilinearMagTest( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, - const armnn::DataLayout dataLayout); +template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> +LayerTestResult<T, 4> ResizeBilinearMagTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>> LayerTestResult<T, 2> Rsqrt2dTestCommon( @@ -1162,26 +1169,6 @@ LayerTestResult<int16_t, 4> ConstantLinearActivationInt16Test( armnn::IWorkloadFactory& workloadFactory, const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); -LayerTestResult<uint8_t, 4> ResizeBilinearNopUint8Test( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); - -LayerTestResult<uint8_t, 4> SimpleResizeBilinearUint8Test( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); - -LayerTestResult<uint8_t, 4> ResizeBilinearSqMinUint8Test( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); - -LayerTestResult<uint8_t, 4> ResizeBilinearMinUint8Test( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); - -LayerTestResult<uint8_t, 4> ResizeBilinearMagUint8Test( - armnn::IWorkloadFactory& workloadFactory, - const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); - LayerTestResult<uint8_t, 4> BatchNormUint8Test( armnn::IWorkloadFactory& workloadFactory, const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager); @@ -2328,3 +2315,540 @@ LayerTestResult<T, 4> SimpleFloorTest( return ret; } + + +template<armnn::DataType ArmnnType, typename T> +LayerTestResult<T, 4> ResizeBilinearNopTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout) +{ + armnn::TensorInfo inputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 4, 4, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 4, 4, dataLayout, ArmnnType); + armnn::TensorInfo outputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 4, 4, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 4, 4, dataLayout, ArmnnType); + if (armnn::IsQuantizedType<T>()) + { + inputTensorInfo.SetQuantizationScale(1.5f); + inputTensorInfo.SetQuantizationOffset(-3); + outputTensorInfo.SetQuantizationScale(1.5f); + outputTensorInfo.SetQuantizationOffset(-3); + } + + std::vector<float> inputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 1, 2, 3, 4, + 2, 3, 4, 5, + 3, 4, 5, 6, + 4, 5, 6, 7 + } + : std::initializer_list<float> + { + 1.0f, 2.0f, 3.0f, 4.0f, + 2.0f, 3.0f, 4.0f, 5.0f, + 3.0f, 4.0f, 5.0f, 6.0f, + 4.0f, 5.0f, 6.0f, 7.0f, + + 1.0f, 2.0f, 3.0f, 4.0f, + 2.0f, 3.0f, 4.0f, 5.0f, + 3.0f, 4.0f, 5.0f, 6.0f, + 4.0f, 5.0f, 6.0f, 7.0f + }; + + const armnn::PermutationVector NCHWToNHWC = { 0, 3, 1, 2 }; + if (dataLayout == armnn::DataLayout::NHWC) + { + std::vector<float> tmp(inputData.size()); + armnnUtils::Permute(inputTensorInfo.GetShape(), NCHWToNHWC, inputData.data(), tmp.data(), sizeof(float)); + inputData = tmp; + } + + auto input = MakeTensor<T, 4>(inputTensorInfo, QuantizedVector<T>(inputTensorInfo.GetQuantizationScale(), + inputTensorInfo.GetQuantizationOffset(), + inputData)); + + LayerTestResult<T, 4> result(outputTensorInfo); + result.outputExpected = input; + + std::unique_ptr<armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo); + std::unique_ptr<armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo); + + armnn::ResizeBilinearQueueDescriptor descriptor; + descriptor.m_Parameters.m_DataLayout = dataLayout; + armnn::WorkloadInfo info; + AddInputToWorkload(descriptor, info, inputTensorInfo, inputHandle.get()); + AddOutputToWorkload(descriptor, info, outputTensorInfo, outputHandle.get()); + + std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateResizeBilinear(descriptor, info); + + inputHandle->Allocate(); + outputHandle->Allocate(); + CopyDataToITensorHandle(inputHandle.get(), &input[0][0][0][0]); + + workload->PostAllocationConfigure(); + workload->Execute(); + + CopyDataFromITensorHandle(&result.output[0][0][0][0], outputHandle.get()); + return result; +} + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::Float32>, 4> +ResizeBilinearNopTest<armnn::DataType::Float32>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::QuantisedAsymm8>, 4> +ResizeBilinearNopTest<armnn::DataType::QuantisedAsymm8>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template<armnn::DataType ArmnnType, typename T> +LayerTestResult<T, 4> SimpleResizeBilinearTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout) +{ + armnn::TensorInfo inputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 2, 2, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 2, 2, dataLayout, ArmnnType); + armnn::TensorInfo outputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 1, 1, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 1, 1, dataLayout, ArmnnType); + + if (armnn::IsQuantizedType<T>()) + { + inputTensorInfo.SetQuantizationScale(0.1567f); + inputTensorInfo.SetQuantizationOffset(1); + outputTensorInfo.SetQuantizationScale(0.1567f); + outputTensorInfo.SetQuantizationOffset(1); + } + + std::vector<float> inputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 1, 255, + 200, 250 + } + : std::initializer_list<float> + { + 1.0f, 255.0f, + 200.0f, 250.0f, + + 250.0f, 200.0f, + 250.0f, 1.0f + }; + + // The 'resize bilinear' operation projects the top-left corner of output texels into the input image, + // then figures out the interpolants and weights. Note this is different to projecting the centre of the + // output texel. Thus, for a input matrix of 2x2, we'll expect the output 1x1 matrix to contain, as + // its single element, the value that was at position (0,0) of the input matrix (rather than an average, + // which we would expect if projecting the centre). + + std::vector<float> outputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 1 + } + : std::initializer_list<float> + { + 1.0f, + + 250.0f + }; + + const armnn::PermutationVector NCHWToNHWC = { 0, 3, 1, 2 }; + if (dataLayout == armnn::DataLayout::NHWC) + { + std::vector<float> tmp(inputData.size()); + armnnUtils::Permute(inputTensorInfo.GetShape(), NCHWToNHWC, inputData.data(), tmp.data(), sizeof(float)); + inputData = tmp; + + std::vector<float> tmp1(outputData.size()); + armnnUtils::Permute(outputTensorInfo.GetShape(), NCHWToNHWC, outputData.data(), tmp1.data(), sizeof(float)); + outputData = tmp1; + } + + auto input = MakeTensor<T, 4>(inputTensorInfo, QuantizedVector<T>(inputTensorInfo.GetQuantizationScale(), + inputTensorInfo.GetQuantizationOffset(), + inputData)); + + LayerTestResult<T, 4> result(outputTensorInfo); + result.outputExpected = MakeTensor<T, 4>(outputTensorInfo, + QuantizedVector<T>(outputTensorInfo.GetQuantizationScale(), + outputTensorInfo.GetQuantizationOffset(), + outputData)); + + std::unique_ptr <armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo); + std::unique_ptr <armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo); + + armnn::ResizeBilinearQueueDescriptor descriptor; + descriptor.m_Parameters.m_DataLayout = dataLayout; + armnn::WorkloadInfo info; + AddInputToWorkload(descriptor, info, inputTensorInfo, inputHandle.get()); + AddOutputToWorkload(descriptor, info, outputTensorInfo, outputHandle.get()); + + std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateResizeBilinear(descriptor, info); + + inputHandle->Allocate(); + outputHandle->Allocate(); + CopyDataToITensorHandle(inputHandle.get(), &input[0][0][0][0]); + + workload->PostAllocationConfigure(); + workload->Execute(); + + CopyDataFromITensorHandle(&result.output[0][0][0][0], outputHandle.get()); + return result; +} + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::Float32>, 4> +SimpleResizeBilinearTest<armnn::DataType::Float32>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::QuantisedAsymm8>, 4> +SimpleResizeBilinearTest<armnn::DataType::QuantisedAsymm8>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template<armnn::DataType ArmnnType, typename T> +LayerTestResult<T, 4> ResizeBilinearSqMinTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout) +{ + armnn::TensorInfo inputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 4, 4, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 4, 4, dataLayout, ArmnnType); + armnn::TensorInfo outputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 2, 2, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 2, 2, dataLayout, ArmnnType); + + if (armnn::IsQuantizedType<T>()) + { + inputTensorInfo.SetQuantizationScale(3.141592f); + inputTensorInfo.SetQuantizationOffset(3); + outputTensorInfo.SetQuantizationScale(3.141592f); + outputTensorInfo.SetQuantizationOffset(3); + } + + std::vector<float> inputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 1, 2, 3, 4, + 2, 3, 4, 5, + 3, 4, 5, 6, + 4, 5, 6, 7 + } + : std::initializer_list<float> + { + 1.0f, 2.0f, 3.0f, 4.0f, + 2.0f, 3.0f, 4.0f, 5.0f, + 3.0f, 4.0f, 5.0f, 6.0f, + 4.0f, 5.0f, 6.0f, 7.0f, + + 7.0f, 6.0f, 5.0f, 4.0f, + 6.0f, 5.0f, 4.0f, 3.0f, + 5.0f, 4.0f, 3.0f, 2.0f, + 4.0f, 3.0f, 2.0f, 1.0f + }; + + std::vector<float> outputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 1, 3, + 3, 5 + } + : std::initializer_list<float> + { + 1.0f, 3.0f, + 3.0f, 5.0f, + + 7.0f, 5.0f, + 5.0f, 3.0f + }; + + const armnn::PermutationVector NCHWToNHWC = { 0, 3, 1, 2 }; + if (dataLayout == armnn::DataLayout::NHWC) + { + std::vector<float> tmp(inputData.size()); + armnnUtils::Permute(inputTensorInfo.GetShape(), NCHWToNHWC, inputData.data(), tmp.data(), sizeof(float)); + inputData = tmp; + + std::vector<float> tmp1(outputData.size()); + armnnUtils::Permute(outputTensorInfo.GetShape(), NCHWToNHWC, outputData.data(), tmp1.data(), sizeof(float)); + outputData = tmp1; + } + + auto input = MakeTensor<T, 4>(inputTensorInfo, QuantizedVector<T>(inputTensorInfo.GetQuantizationScale(), + inputTensorInfo.GetQuantizationOffset(), + inputData)); + + LayerTestResult<T, 4> result(outputTensorInfo); + result.outputExpected = MakeTensor<T, 4>(outputTensorInfo, + QuantizedVector<T>(outputTensorInfo.GetQuantizationScale(), + outputTensorInfo.GetQuantizationOffset(), + outputData)); + + std::unique_ptr <armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo); + std::unique_ptr <armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo); + + armnn::ResizeBilinearQueueDescriptor descriptor; + descriptor.m_Parameters.m_DataLayout = dataLayout; + armnn::WorkloadInfo info; + AddInputToWorkload(descriptor, info, inputTensorInfo, inputHandle.get()); + AddOutputToWorkload(descriptor, info, outputTensorInfo, outputHandle.get()); + + std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateResizeBilinear(descriptor, info); + + inputHandle->Allocate(); + outputHandle->Allocate(); + CopyDataToITensorHandle(inputHandle.get(), &input[0][0][0][0]); + + workload->PostAllocationConfigure(); + workload->Execute(); + + CopyDataFromITensorHandle(&result.output[0][0][0][0], outputHandle.get()); + return result; +} + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::Float32>, 4> +ResizeBilinearSqMinTest<armnn::DataType::Float32>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::QuantisedAsymm8>, 4> +ResizeBilinearSqMinTest<armnn::DataType::QuantisedAsymm8>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template<armnn::DataType ArmnnType, typename T> +LayerTestResult<T, 4> ResizeBilinearMinTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout) +{ + armnn::TensorInfo inputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 2, 3, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 3, 5, dataLayout, ArmnnType); + armnn::TensorInfo outputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 1, 2, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 2, 3, dataLayout, ArmnnType); + + if (armnn::IsQuantizedType<T>()) + { + inputTensorInfo.SetQuantizationScale(1.5f); + inputTensorInfo.SetQuantizationOffset(-1); + outputTensorInfo.SetQuantizationScale(1.5f); + outputTensorInfo.SetQuantizationOffset(-1); + } + + std::vector<float> inputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 3.0f, 4.5f, 6.0f, // 1, 2, 3, : Expected quantised values + 9.0f, 13.5f, 21.0f // 5, 8, 13 + } + : std::initializer_list<float> + { + 1.0f, 2.0f, 3.0f, 5.0f, 8.0f, + 13.0f, 21.0f, 34.0f, 55.0f, 89.0f, + 144.0f, 233.0f, 377.0f, 610.0f, 987.0f, + + 987.0f, 610.0f, 377.0f, 233.0f, 144.0f, + 89.0f, 55.0f, 34.0f, 21.0f, 13.0f, + 8.0f, 5.0f, 3.0f, 2.0f, 1.0f + }; + + std::vector<float> outputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 3.0f, 5.25f // 1, 3 + } + : std::initializer_list<float> + { + 1.0f, 2.6666f, 6.00f, + 78.5f, 179.3333f, 401.00f, + + 987.0f, 454.6670f, 203.33f, + 48.5f, 22.3333f, 10.00f + }; + + const armnn::PermutationVector NCHWToNHWC = { 0, 3, 1, 2 }; + if (dataLayout == armnn::DataLayout::NHWC) + { + std::vector<float> tmp(inputData.size()); + armnnUtils::Permute(inputTensorInfo.GetShape(), NCHWToNHWC, inputData.data(), tmp.data(), sizeof(float)); + inputData = tmp; + + std::vector<float> tmp1(outputData.size()); + armnnUtils::Permute(outputTensorInfo.GetShape(), NCHWToNHWC, outputData.data(), tmp1.data(), sizeof(float)); + outputData = tmp1; + } + + auto input = MakeTensor<T, 4>(inputTensorInfo, QuantizedVector<T>(inputTensorInfo.GetQuantizationScale(), + inputTensorInfo.GetQuantizationOffset(), + inputData)); + + LayerTestResult<T, 4> result(outputTensorInfo); + result.outputExpected = MakeTensor<T, 4>(outputTensorInfo, + QuantizedVector<T>(outputTensorInfo.GetQuantizationScale(), + outputTensorInfo.GetQuantizationOffset(), + outputData)); + + std::unique_ptr<armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo); + std::unique_ptr<armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo); + + armnn::ResizeBilinearQueueDescriptor descriptor; + descriptor.m_Parameters.m_DataLayout = dataLayout; + armnn::WorkloadInfo info; + AddInputToWorkload(descriptor, info, inputTensorInfo, inputHandle.get()); + AddOutputToWorkload(descriptor, info, outputTensorInfo, outputHandle.get()); + + std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateResizeBilinear(descriptor, info); + + inputHandle->Allocate(); + outputHandle->Allocate(); + CopyDataToITensorHandle(inputHandle.get(), &input[0][0][0][0]); + + workload->PostAllocationConfigure(); + workload->Execute(); + + CopyDataFromITensorHandle(&result.output[0][0][0][0], outputHandle.get()); + return result; +} + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::Float32>, 4> +ResizeBilinearMinTest<armnn::DataType::Float32>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::QuantisedAsymm8>, 4> +ResizeBilinearMinTest<armnn::DataType::QuantisedAsymm8>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template<armnn::DataType ArmnnType, typename T> +LayerTestResult<T, 4> ResizeBilinearMagTest( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout) +{ + armnn::TensorInfo inputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 3, 2, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 3, 2, dataLayout, ArmnnType); + armnn::TensorInfo outputTensorInfo = armnn::IsQuantizedType<T>() + ? armnnUtils::GetTensorInfo(1, 1, 3, 5, dataLayout, ArmnnType) + : armnnUtils::GetTensorInfo(1, 2, 3, 5, dataLayout, ArmnnType); + + if (armnn::IsQuantizedType<T>()) + { + inputTensorInfo.SetQuantizationScale(0.010765f); + inputTensorInfo.SetQuantizationOffset(7); + outputTensorInfo.SetQuantizationScale(0.010132f); + outputTensorInfo.SetQuantizationOffset(-18); + } + + std::vector<float> inputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 0.183005f, 2.379065f, // 24, 228, : Expected quantised values + 1.05497f, 1.302565f, // 105, 128, + 2.400595f, 0.68896f // 230, 71 + } + : std::initializer_list<float> + { + 1.0f, 2.0f, + 13.0f, 21.0f, + 144.0f, 233.0f, + + 233.0f, 144.0f, + 21.0f, 13.0f, + 2.0f, 1.0f + }; + std::vector<float> outputData = armnn::IsQuantizedType<T>() + ? std::initializer_list<float> + { + 0.18300501f, 1.06142902f, 1.93985295f, 2.37906504f, 2.37906504f, + 1.05497003f, 1.15400803f, 1.25304604f, 1.30256498f, 1.30256498f, + 2.40059495f, 1.71594095f, 1.03128707f, 0.68896002f, 0.68896002f + // 0, 87, 173, 217, 217, : Expected quantised values + // 86, 96, 106, 111, 111, + // 219, 151, 84, 50, 50 + } + : std::initializer_list<float> + { + 1.0f, 1.4f, 1.8f, 2.0f, 2.0f, + 13.0f, 16.2f, 19.4f, 21.0f, 21.0f, + 144.0f, 179.6f, 215.2f, 233.0f, 233.0f, + + 233.0f, 197.4f, 161.8f, 144.0f, 144.0f, + 21.0f, 17.8f, 14.6f, 13.0f, 13.0f, + 2.0f, 1.6f, 1.2f, 1.0f, 1.0f + }; + + const armnn::PermutationVector NCHWToNHWC = { 0, 3, 1, 2 }; + if (dataLayout == armnn::DataLayout::NHWC) + { + std::vector<float> tmp(inputData.size()); + armnnUtils::Permute(inputTensorInfo.GetShape(), NCHWToNHWC, inputData.data(), tmp.data(), sizeof(float)); + inputData = tmp; + + std::vector<float> tmp1(outputData.size()); + armnnUtils::Permute(outputTensorInfo.GetShape(), NCHWToNHWC, outputData.data(), tmp1.data(), sizeof(float)); + outputData = tmp1; + } + + auto input = MakeTensor<T, 4>(inputTensorInfo, QuantizedVector<T>(inputTensorInfo.GetQuantizationScale(), + inputTensorInfo.GetQuantizationOffset(), + inputData)); + + LayerTestResult<T, 4> result(outputTensorInfo); + result.outputExpected = MakeTensor<T, 4>(outputTensorInfo, + QuantizedVector<T>(outputTensorInfo.GetQuantizationScale(), + outputTensorInfo.GetQuantizationOffset(), + outputData)); + + std::unique_ptr <armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo); + std::unique_ptr <armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo); + + armnn::ResizeBilinearQueueDescriptor descriptor; + descriptor.m_Parameters.m_DataLayout = dataLayout; + armnn::WorkloadInfo info; + AddInputToWorkload(descriptor, info, inputTensorInfo, inputHandle.get()); + AddOutputToWorkload(descriptor, info, outputTensorInfo, outputHandle.get()); + + std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateResizeBilinear(descriptor, info); + + inputHandle->Allocate(); + outputHandle->Allocate(); + CopyDataToITensorHandle(inputHandle.get(), &input[0][0][0][0]); + + workload->PostAllocationConfigure(); + workload->Execute(); + + CopyDataFromITensorHandle(&result.output[0][0][0][0], outputHandle.get()); + return result; +} + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::Float32>, 4> +ResizeBilinearMagTest<armnn::DataType::Float32>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout); + +template LayerTestResult<typename armnn::ResolveType<armnn::DataType::QuantisedAsymm8>, 4> +ResizeBilinearMagTest<armnn::DataType::QuantisedAsymm8>( + armnn::IWorkloadFactory& workloadFactory, + const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager, + const armnn::DataLayout dataLayout);
\ No newline at end of file |