From ba84dce20fcf0f146a7c1da7125bd9486a2979fb Mon Sep 17 00:00:00 2001 From: Wook Song Date: Mon, 13 Jan 2020 15:39:39 +0900 Subject: [PATCH] [Filter/OpenVino] Keep consistency in the private member variable names In order to improve the readability of code, this patch makes the names of the private member variables use a prefix, '_', to keep consistency. Signed-off-by: Wook Song --- .../tensor_filter/tensor_filter_openvino.cc | 32 +++++++++++----------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/ext/nnstreamer/tensor_filter/tensor_filter_openvino.cc b/ext/nnstreamer/tensor_filter/tensor_filter_openvino.cc index 67f5671..f405fab 100644 --- a/ext/nnstreamer/tensor_filter/tensor_filter_openvino.cc +++ b/ext/nnstreamer/tensor_filter/tensor_filter_openvino.cc @@ -69,7 +69,7 @@ public: // TODO: Need to support other acceleration devices int loadModel (accl_hw hw); bool isModelLoaded () { - return isLoaded; + return _isLoaded; } int getInputTensorDim (GstTensorsInfo * info); @@ -96,10 +96,10 @@ private: InferenceEngine::InferRequest _inferRequest; static std::map _nnsAcclHwToOVDevMap; - std::string pathModelXml; - std::string pathModelBin; - bool isLoaded; - accl_hw hw; + std::string _pathModelXml; + std::string _pathModelBin; + bool _isLoaded; + accl_hw _hw; }; std::map TensorFilterOpenvino::_nnsAcclHwToOVDevMap = { @@ -228,7 +228,7 @@ TensorFilterOpenvino::isAcclDevSupported (std::vector &devsVector, std::string TensorFilterOpenvino::getPathModelXml () { - return this->pathModelXml; + return this->_pathModelXml; } /** @@ -238,7 +238,7 @@ TensorFilterOpenvino::getPathModelXml () std::string TensorFilterOpenvino::getPathModelBin () { - return this->pathModelBin; + return this->_pathModelBin; } /** @@ -250,15 +250,15 @@ TensorFilterOpenvino::getPathModelBin () TensorFilterOpenvino::TensorFilterOpenvino (std::string pathModelXml, std::string pathModelBin) { - this->pathModelXml = pathModelXml; - this->pathModelBin = pathModelBin; - (this->_networkReaderCNN).ReadNetwork (this->pathModelXml); - (this->_networkReaderCNN).ReadWeights (this->pathModelBin); + this->_pathModelXml = pathModelXml; + this->_pathModelBin = pathModelBin; + (this->_networkReaderCNN).ReadNetwork (this->_pathModelXml); + (this->_networkReaderCNN).ReadWeights (this->_pathModelBin); this->_networkCNN = (this->_networkReaderCNN).getNetwork (); this->_inputsDataMap = (this->_networkCNN).getInputsInfo (); this->_outputsDataMap = (this->_networkCNN).getOutputsInfo (); - this->isLoaded = false; - this->hw = ACCL_NONE; + this->_isLoaded = false; + this->_hw = ACCL_NONE; } /** @@ -282,7 +282,7 @@ TensorFilterOpenvino::loadModel (accl_hw hw) std::vector strVector; std::vector::iterator strVectorIter; - if (this->isLoaded) { + if (this->_isLoaded) { // TODO: Can OpenVino support to replace the loaded model with a new one? g_critical ("The model file is already loaded onto the device."); return RetEBusy; @@ -311,8 +311,8 @@ TensorFilterOpenvino::loadModel (accl_hw hw) /** TODO: Catch the IE exception */ this->_executableNet = this->_ieCore.LoadNetwork (this->_networkCNN, _nnsAcclHwToOVDevMap[hw]); - this->hw = hw; - this->isLoaded = true; + this->_hw = hw; + this->_isLoaded = true; this->_inferRequest = this->_executableNet.CreateInferRequest (); return RetSuccess; -- 2.7.4