Find your GPU and add its respective Compute Capability to your project's CMakeLists.txt file, under the CMAKE_CUDA_ARCHITECTURES
variable, like so:
SET(CMAKE_CUDA_ARCHITECTURES 61)
Remember to remove the decimal point in the Compute Capability version number!