我按照老师的知乎配置好caffe后，根据https://blog.csdn.net/u012986684/article/details/72469542进行build的编译。
因为有我在使用create_list.sh 的时候会报…/build/tools/get_image_size: 没有那个文件或目这个错误

但是，我对build cmake… 和 make all 的时候会报
Building NVCC (Device) object src/caffe/CMakeFiles/cuda_compile.dir/layers/cuda_compile_generated_embed_layer.cu.o
nvcc fatal : Unsupported gpu architecture 'compute_75’
CMake Error at cuda_compile_generated_embed_layer.cu.o.cmake:207 (message):
Error generating
/home/hyb/caffe/build/src/caffe/CMakeFiles/cuda_compile.dir/layers/./cuda_compile_generated_embed_layer.cu.o

src/caffe/CMakeFiles/caffe.dir/build.make:469: recipe for target ‘src/caffe/CMakeFiles/cuda_compile.dir/layers/cuda_compile_generated_embed_layer.cu.o’ failed
make[2]: *** [src/caffe/CMakeFiles/cuda_compile.dir/layers/cuda_compile_generated_embed_layer.cu.o] Error 1
CMakeFiles/Makefile2:304: recipe for target ‘src/caffe/CMakeFiles/caffe.dir/all’ failed
make[1]: *** [src/caffe/CMakeFiles/caffe.dir/all] Error 2
Makefile:127: recipe for target ‘all’ failed
make: *** [all] Error 2

我认为：nvcc fatal : Unsupported gpu architecture 'compute_75’是这里出错了，我百度了很久，没找了答案，我不知道是怎么改，我cuda和cudnn是和Python3+TensorFlow人脸智能小程序的配置是一样的，应该不会是这里的问题吧？？