Zhiheng Li

You can also add multiprocessing.set_start_method('spawn') following `if __name__ == '__main__':`

I use g++-4.8 to compile the source file but with the same error: `fatal error: cuda/cuda_config.h: No such file or directory #include "cuda/cuda_config.h"` I install tensorflow via pip. I use...

It works! Thank you!

I encounter the same problem (undefined symbol error) as @GitHubShily said before. I've also reinstalled my g++-4.8 but it doesn't help.

It works! Thank you! @cotrane

gcc-4.8 works as well. Adding -ccbin flag to nvcc is the key issue.