RuntimeError: Error compiling objects for extension about rdpnet HOT 5 CLOSED

Yes, this error is usually caused by an incompatible CUDA version.
As default, this repo uses system CUDA library to compile the cuda source code. So you need to prepare a same version CUDA library.

from rdpnet.

But I use the CUDA10.2 Library of the system to run setup.py, but I can't compile it. Do I have to call the CUDA Library of the system back to 9.0?

from rdpnet.

But I use the CUDA10.2 Library of the system to run setup.py, but I can't compile it. Do I have to call the CUDA Library of the system back to 9.0?

Yes. This issue is normal and can be classified to the incompatible version error.
I recommend to use my default env "pytorch==1.1 torchvision==0.2.2 cudatoolkit=9.0" and make sure the version of the CUDA library (used to compile) is CUDA9.0.

After installing CUDA9.0, you can consider to use my command to conveniently activate CUDA9.0

### Write the following code to ~/.bashrc. Use `source ~/.bashrc` to activate it
alias cu102='sudo rm /usr/local/cuda; sudo ln -s /usr/local/cuda-10.2 /usr/local/cuda; echo enable cuda10.2!; nvcc -V'
alias cu101='sudo rm /usr/local/cuda; sudo ln -s /usr/local/cuda-10.1 /usr/local/cuda; echo enable cuda10.1!; nvcc -V'
alias cu92='sudo rm /usr/local/cuda; sudo ln -s /usr/local/cuda-9.2 /usr/local/cuda; echo enable cuda9.2!; nvcc -V'
alias cu90='sudo rm /usr/local/cuda; sudo ln -s /usr/local/cuda-9.0 /usr/local/cuda; echo enable cuda9.0!; nvcc -V'

### Use the following code in the terminal to activate CUDA9.0
cu90

Hope this answer will help you a lot.

from rdpnet.

If you have further questions, feel free to add my wechat "wyh-hys" with your current university and current position~

from rdpnet.

OK, thank you for your answer.

from rdpnet.