Hi,

I have both /usr/local/cuda-10.0 and /usr/local/cuda-9.0 directories.

When I do `ls`

under /usr/local/cuda-10.0/lib64 directory, I get

libaccinj64.so libnppc.so libnppist_static.a

libaccinj64.so.10.1 libnppc.so.10 libnppisu.so

libaccinj64.so.10.1.168 libnppc.so.10.1.168 libnppisu.so.10

libcudadevrt.a libnppc_static.a libnppisu.so.10.1.168

libcudart.so libnppial.so libnppisu_static.a

libcudart.so.10.1 libnppial.so.10 libnppitc.so

libcudart.so.10.1.168 libnppial.so.10.1.168 libnppitc.so.10

libcudart_static.a libnppial_static.a libnppitc.so.10.1.168

libcufft.so libnppicc.so libnppitc_static.a

libcufft.so.10 libnppicc.so.10 libnpps.so

libcufft.so.10.1.168 libnppicc.so.10.1.168 libnpps.so.10

libcufft_static.a libnppicc_static.a libnpps.so.10.1.168

libcufft_static_nocallback.a libnppicom.so libnpps_static.a

…

I think I should have libxxx.so.10.0.xxx files and libxxx.so.10.1 links.

and I thought I installed cudnn but can’t find those files there.

I need to run tensorflow (which expects CUDA 10.0), do I have re-install cuda-10.0?

(I tested with fashion_mnist.py and it seems running ok with GPUs but complains it cannot find libcudart.so.10.0 libcublas.so.10.0 libcufft.so.10.0 libcurand.so.10.0 libcusolver.so.10.0 libcusparse.so.10.0 files. (so tensorflow is looking for cuda 10.0)

Is this normal? or something is messed?

I’m considering re-installing cuda-10.0 and cudnn version for that.

Please give me some advices. Thank you!