diff options
Diffstat (limited to 'src/armnn/test/optimizations/FoldPadTests.cpp')
| -rw-r--r-- | src/armnn/test/optimizations/FoldPadTests.cpp | 67 |
1 files changed, 34 insertions, 33 deletions
diff --git a/src/armnn/test/optimizations/FoldPadTests.cpp b/src/armnn/test/optimizations/FoldPadTests.cpp index 20cfab1cb7..7b4ac4170f 100644 --- a/src/armnn/test/optimizations/FoldPadTests.cpp +++ b/src/armnn/test/optimizations/FoldPadTests.cpp @@ -6,15 +6,16 @@ #include "LayersFwd.hpp" #include <Network.hpp> #include <test/TestUtils.hpp> -#include <boost/test/unit_test.hpp> +#include <doctest/doctest.h> #include <backendsCommon/TensorHandle.hpp> #include <Optimizer.hpp> -BOOST_AUTO_TEST_SUITE(Optimizer) +TEST_SUITE("Optimizer") +{ using namespace armnn; using namespace armnn::optimizations; -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConvolution2dLayer) +TEST_CASE("FoldPadLayerIntoConvolution2dLayer") { Graph graph; const unsigned int inputShape[] = {1, 2, 2, 3}; @@ -67,7 +68,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConvolution2dLayer) (conv2dLayerParams.m_BiasEnabled == false) && (conv2dLayerParams.m_DataLayout == DataLayout::NHWC); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimpleConv2d, @@ -85,13 +86,13 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConvolution2dLayer) (conv2dLayerParams.m_BiasEnabled == false) && (conv2dLayerParams.m_DataLayout == DataLayout::NHWC); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, checkPadFoldedIntoConv2d, &IsLayerOfType<OutputLayer>)); } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConvolution2dLayer) +TEST_CASE("FoldPadLayerIntoDepthwiseConvolution2dLayer") { Graph graph; const unsigned int inputShape[] = {1, 2, 2, 3}; @@ -146,7 +147,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConvolution2dLayer) (depthwiseConv2dLayerParams.m_DataLayout == DataLayout::NHWC); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimpleDepthwiseConv2d, @@ -166,13 +167,13 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConvolution2dLayer) (depthwiseConv2dLayerParams.m_DataLayout == DataLayout::NHWC); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, checkPadFoldedIntoDepthwiseConv2d, &IsLayerOfType<OutputLayer>)); } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer) +TEST_CASE("FoldPadLayerIntoPooling2dLayer") { Graph graph; const unsigned int inputShape[] = {1, 2, 2, 3}; @@ -218,7 +219,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer) (pool2dLayer->GetParameters() == pooling2dDescriptor); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -248,13 +249,13 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer) (pool2dLayerParams.m_PadBottom == 1) && (pool2dLayerParams.m_PaddingMethod == PaddingMethod::IgnoreValue); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, checkPadFoldedIntoPool2d, &IsLayerOfType<OutputLayer>)); } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2d_PadWithMultipleOutputsShouldNotBeOptimized) +TEST_CASE("FoldPadLayerIntoPooling2d_PadWithMultipleOutputsShouldNotBeOptimized") { // In this test case we'll setup a pad layer with two outputs. One goes to a polling layers and the other // goes to an output layer. FoldPadLayerIntoPooling2d should not optimize this graph as it uses the @@ -308,7 +309,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2d_PadWithMultipleOutputsShouldNotBe }; // Initial sequence. - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -318,7 +319,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2d_PadWithMultipleOutputsShouldNotBe armnn::Optimizer::Pass(graph, MakeOptimizations(FoldPadIntoPooling2d())); // The network should not change. - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -326,7 +327,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2d_PadWithMultipleOutputsShouldNotBe &IsLayerOfType<OutputLayer>)); } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_PoolingLayerWithExcludePaddingShouldNotTakeMorePadding) +TEST_CASE("FoldPadLayerIntoPooling2dLayer_PoolingLayerWithExcludePaddingShouldNotTakeMorePadding") { // In this test setup input, Pad layer, Pooling layer that includes padding, output layer. The optimization // should not work as the pooling layer already includes and existing pad and specifies PaddingMethod::Exclude. @@ -380,7 +381,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_PoolingLayerWithExcludePaddi (pool2dLayer->GetParameters() == pooling2dDescriptor); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -389,14 +390,14 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_PoolingLayerWithExcludePaddi armnn::Optimizer::Pass(graph, MakeOptimizations(FoldPadIntoPooling2d())); // The optimization should not have modified the graph. - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, &IsLayerOfType<OutputLayer>)); } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_MaxPoolingLayerWithLargePadValueShouldNotBeFolded) +TEST_CASE("FoldPadLayerIntoPooling2dLayer_MaxPoolingLayerWithLargePadValueShouldNotBeFolded") { // In this test setup input, Pad layer with a large pad value, Max Pooling layer, output layer. The optimization // should not work as the pad value will modify the result of the max pooling layer. @@ -447,7 +448,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_MaxPoolingLayerWithLargePadV (pool2dLayer->GetParameters() == pooling2dDescriptor); }; - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -456,7 +457,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_MaxPoolingLayerWithLargePadV armnn::Optimizer::Pass(graph, MakeOptimizations(FoldPadIntoPooling2d())); // The optimization should not have modified the graph. - BOOST_TEST(CheckSequence(graph.cbegin(), graph.cend(), + CHECK(CheckSequence(graph.cbegin(), graph.cend(), &IsLayerOfType<InputLayer>, &IsLayerOfType<PadLayer>, checkSimplePool2d, @@ -464,7 +465,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_MaxPoolingLayerWithLargePadV } #if defined(ARMNNREF_ENABLED) -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWithoutOptimization) +TEST_CASE("FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWithoutOptimization") { // The idea of this test to run a simple pad+pool2d network twice. Once // with FoldPadLayerIntoPooling2dLayer enabled and a second time with it @@ -523,7 +524,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWitho IOptimizedNetworkPtr optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); // Load network into runtime NetworkId networkIdentifier; - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); InputTensors inputTensors{{0, ConstTensor(run->GetInputTensorInfo(networkIdentifier, 0), inputData.data())}}; @@ -544,7 +545,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWitho // Optimize and load and execute it a second time. optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); std::vector<float> goldenData(32, 0.0f); std::vector<float> padOutputData(72, 0.0f); OutputTensors goldenTensors{{0, Tensor(outputInfo, goldenData.data())}, @@ -552,7 +553,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWitho run->EnqueueWorkload(networkIdentifier, inputTensors, goldenTensors); // Now we can compare goldenData against optimizedData. They should be the same. - BOOST_TEST(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); + CHECK(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); } catch (const std::exception& e) { @@ -561,7 +562,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoPooling2dLayer_ExecuteInferenceWithAndWitho } } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutOptimization) +TEST_CASE("FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutOptimization") { // The idea of this test to run a simple pad+conv2d network twice. Once // with FoldPadLayerIntoConv2dLayer enabled and a second time with it @@ -641,7 +642,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutO IOptimizedNetworkPtr optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); // Load network into runtime NetworkId networkIdentifier; - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); InputTensors inputTensors{{0, ConstTensor(run->GetInputTensorInfo(networkIdentifier, 0), inputData.data())}}; @@ -662,7 +663,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutO // Optimize and load and execute it a second time. optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); std::vector<float> goldenData(100, 0.0f); std::vector<float> padOutputData(108, 0.0f); OutputTensors goldenTensors{{0, Tensor(outputInfo, goldenData.data())}, @@ -670,7 +671,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutO run->EnqueueWorkload(networkIdentifier, inputTensors, goldenTensors); // Now we can compare goldenData against optimizedData. They should be the same. - BOOST_TEST(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); + CHECK(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); } catch (const std::exception& e) { @@ -679,7 +680,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoConv2dLayer_ExecuteInferenceWithAndWithoutO } } -BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAndWithoutOptimization) +TEST_CASE("FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAndWithoutOptimization") { // The idea of this test to run a simple pad+depthwiseconv2d network twice. Once // with FoldPadLayerIntoDeptwiseConv2dLayer enabled and a second time with it @@ -759,7 +760,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAn IOptimizedNetworkPtr optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); // Load network into runtime NetworkId networkIdentifier; - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); InputTensors inputTensors{{0, ConstTensor(run->GetInputTensorInfo(networkIdentifier, 0), inputData.data())}}; @@ -780,7 +781,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAn // Optimize and load and execute it a second time. optimizedNetwork = Optimize(*network, {Compute::CpuRef}, run->GetDeviceSpec()); - BOOST_TEST(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); + CHECK(run->LoadNetwork(networkIdentifier, std::move(optimizedNetwork)) == Status::Success); std::vector<float> goldenData(300, 0.0f); std::vector<float> padOutputData(108, 0.0f); OutputTensors goldenTensors{{0, Tensor(outputInfo, goldenData.data())}, @@ -788,7 +789,7 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAn run->EnqueueWorkload(networkIdentifier, inputTensors, goldenTensors); // Now we can compare goldenData against optimizedData. They should be the same. - BOOST_TEST(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); + CHECK(std::equal(goldenData.begin(), goldenData.end(), optimizedData.begin())); } catch (const std::exception& e) { @@ -798,4 +799,4 @@ BOOST_AUTO_TEST_CASE(FoldPadLayerIntoDepthwiseConv2dLayer_ExecuteInferenceWithAn } #endif -BOOST_AUTO_TEST_SUITE_END()
\ No newline at end of file +}
\ No newline at end of file |