• Hardware Platform - Jetson Orin Nano (Super Mode)
• DeepStream Version - 7.1
• JetPack Version (valid for Jetson only) - 6.2
• TensorRT Version - 10.3.0.30-1+cuda12.5
• Issue Type - Bug
• How to reproduce the issue ? - Follow the Build nvOCDR library , when used make command, I got the below issue
~/onnx_models/NVIDIA-Optical-Character-Detection-and-Recognition-Solution$ make
if [ ! -d ./chobj ]; then mkdir -p ./chobj; fi
g++ -MM -MF chobj/TRTEngine.d -MP -MT chobj/TRTEngine.o -std=c++11 -I./include -I/usr/include/x86_64-linux-gnu/ -I./src -I/usr/local/cuda/include -I/usr/include/opencv4/ src/TRTEngine.cpp
Compiling: src/TRTEngine.cpp
g++ -std=c++11 -I./include -I/usr/include/x86_64-linux-gnu/ -I./src -I/usr/local/cuda/include -I/usr/include/opencv4/ -fPIC -c -o chobj/TRTEngine.o src/TRTEngine.cpp
src/TRTEngine.cpp: In member function ‘size_t nvocdr::TRTEngine::getTrtIoTensorDtypeSizeByName(std::string&)’:
src/TRTEngine.cpp:187:33: error: ‘class nvinfer1::ICudaEngine’ has no member named ‘getBindingIndex’
187 | int bindingIndex = mEngine->getBindingIndex(tensorName.c_str());
| ^~~~~~~~~~~~~~~
src/TRTEngine.cpp:192:44: error: ‘class nvinfer1::ICudaEngine’ has no member named ‘getBindingDataType’; did you mean ‘getTensorDataType’?
192 | nvinfer1::DataType dataType = mEngine->getBindingDataType(bindingIndex);
| ^~~~~~~~~~~~~~~~~~
| getTensorDataType
I made sure I followed all the steps without error.
Note: When I rolled back to previous commit of NVIDIA_AI_IOT/nvOCDR I can able to run the nvOCDR Sample using gst-launch pipeline.
Kindly tell me how to resolve this.
Thanks