diff options
Diffstat (limited to 'src/armnn/test/CreateWorkload.hpp')
| -rw-r--r-- | src/armnn/test/CreateWorkload.hpp | 24 |
1 files changed, 12 insertions, 12 deletions
diff --git a/src/armnn/test/CreateWorkload.hpp b/src/armnn/test/CreateWorkload.hpp index f6928f858f..02ce12a304 100644 --- a/src/armnn/test/CreateWorkload.hpp +++ b/src/armnn/test/CreateWorkload.hpp @@ -399,12 +399,12 @@ std::unique_ptr<QuantizedLstmWorkload> CreateQuantizedLstmWorkloadTest(armnn::IW // Weights and bias tensor and quantization info armnn::TensorInfo inputWeightsInfo({outputSize, inputSize}, - armnn::DataType::QuantisedAsymm8, + armnn::DataType::QAsymmU8, weightsScale, weightsOffset); armnn::TensorInfo recurrentWeightsInfo({outputSize, outputSize}, - armnn::DataType::QuantisedAsymm8, + armnn::DataType::QAsymmU8, weightsScale, weightsOffset); @@ -463,17 +463,17 @@ std::unique_ptr<QuantizedLstmWorkload> CreateQuantizedLstmWorkloadTest(armnn::IW // Input/output tensor info and quantization info armnn::TensorInfo inputInfo({numBatches , inputSize}, - armnn::DataType::QuantisedAsymm8, + armnn::DataType::QAsymmU8, inputOutputScale, inputOutputOffset); armnn::TensorInfo cellStateInfo({numBatches , outputSize}, - armnn::DataType::QuantisedSymm16, + armnn::DataType::QSymmS16, cellStateScale, cellStateOffset); armnn::TensorInfo outputStateInfo({numBatches , outputSize}, - armnn::DataType::QuantisedAsymm8, + armnn::DataType::QAsymmU8, inputOutputScale, inputOutputOffset); @@ -530,8 +530,8 @@ std::unique_ptr<Convolution2dWorkload> CreateDirectConvolution2dWorkloadTest(arm Convolution2dLayer* const layer = graph.AddLayer<Convolution2dLayer>(layerDesc, "layer"); - float inputsQScale = DataType == armnn::DataType::QuantisedAsymm8 ? 1.0f : 0.0; - float outputQScale = DataType == armnn::DataType::QuantisedAsymm8 ? 2.0f : 0.0; + float inputsQScale = DataType == armnn::DataType::QAsymmU8 ? 1.0f : 0.0; + float outputQScale = DataType == armnn::DataType::QAsymmU8 ? 2.0f : 0.0; layer->m_Weight = std::make_unique<ScopedCpuTensorHandle>(TensorInfo({ 2, 3, 3, 3 }, DataType, inputsQScale)); layer->m_Bias = std::make_unique<ScopedCpuTensorHandle> @@ -637,8 +637,8 @@ std::unique_ptr<FullyConnectedWorkload> CreateFullyConnectedWorkloadTest(armnn:: FullyConnectedLayer* const layer = graph.AddLayer<FullyConnectedLayer>(layerDesc, "layer"); - float inputsQScale = DataType == armnn::DataType::QuantisedAsymm8 ? 1.0f : 0.0; - float outputQScale = DataType == armnn::DataType::QuantisedAsymm8 ? 2.0f : 0.0; + float inputsQScale = DataType == armnn::DataType::QAsymmU8 ? 1.0f : 0.0; + float outputQScale = DataType == armnn::DataType::QAsymmU8 ? 2.0f : 0.0; layer->m_Weight = std::make_unique<ScopedCpuTensorHandle>(TensorInfo({7, 20}, DataType, inputsQScale, 0)); layer->m_Bias = std::make_unique<ScopedCpuTensorHandle>(TensorInfo({7}, GetBiasDataType(DataType), inputsQScale)); @@ -1361,7 +1361,7 @@ std::pair<armnn::IOptimizedNetworkPtr, std::unique_ptr<PreCompiledWorkload>> Cre if (biasEnabled) { - constexpr armnn::DataType biasDataType = ( dataType == armnn::DataType::QuantisedAsymm8) ? + constexpr armnn::DataType biasDataType = ( dataType == armnn::DataType::QAsymmU8) ? armnn::DataType::Signed32 : armnn::DataType::Float32; TensorInfo biasTensorInfo(TensorShape({16}), biasDataType, 0.9f * 0.9f, 0); @@ -1396,14 +1396,14 @@ std::pair<armnn::IOptimizedNetworkPtr, std::unique_ptr<PreCompiledWorkload>> Cre // set the tensors in the network (NHWC format) TensorInfo inputTensorInfo(TensorShape({ 1, 16, 16, 16 }), dataType); - if (dataType == armnn::DataType::QuantisedAsymm8) + if (dataType == armnn::DataType::QAsymmU8) { inputTensorInfo.SetQuantizationOffset(0); inputTensorInfo.SetQuantizationScale(0.9f); } TensorInfo outputTensorInfo(TensorShape({1, 16, 16, 16}), dataType); - if (dataType == armnn::DataType::QuantisedAsymm8) + if (dataType == armnn::DataType::QAsymmU8) { outputTensorInfo.SetQuantizationOffset(0); outputTensorInfo.SetQuantizationScale(0.9f); |