Top Answers

1. Unsupported gpu architecture compute_30 on a CUDA 5 capable gpu
   27 votes
2. How to destroy a stream that was created on a specific device?
   6 votes
3. How to deal with "DNN module was not built with CUDA backend; switching to CPU" warning in C++?
   4 votes
4. CUDA optimization for a vector tensor product using a custom kernel or CUBLAS
   3 votes
5. Installation issue of CUDA and cuDNN on Windows
   2 votes