❓ Q&A for Building TensorRT Engines
1. Error: trtexec: not found
Solution: Try the following command
source trt/activate.sh
This command is used to configure the environment variables required to build the TensorRT Engine.
2. Error: Unable to open library: libnvinfer_plugin.so.9 due to libcublas.so.11: cannot open shared object file
Solution: This error occurs because the libnvinfer_plugin.so.9 library depends on the libcublas.so.11 library.
However, you may use a different version of cuda library (see /usr/local/cuda), such as cuda-12. It is recommended
to use cuda-11 to build the TensorRT Engine.
3. The size of the engine file generated by each build is different
Explanation: During the process of building the engine, the building program will automatically select the optimal kernel and other parameters. Therefore, the size of the engine file generated by each build may be different. Therefore, please ensure that the GPU is idle during the build to ensure that the building program correctly selects the optimal kernel.