Nvcc fatal : Value 'sm_75' is not defined for option 'gpu-architecture' CUDA 10.0

Hi,
I’m trying to run the StyleGAN2 sample using CUDA 10.0 on RTX 2080.

Here’s the log output:
nvcc “/code/generative_models/style_gan2_play/venv/lib/python3.6/site-packages/tensorflow_core/python/_pywrap_tensorflow_internal.so” --compiler-options ‘-fPIC -D_GLIBCXX_USE_CXX11_ABI=0’ --gpu-architecture=sm_75 --use_fast_math --disable-warnings --include-path “/code/generative_models/style_gan2_play/venv/lib/python3.6/site-packages/tensorflow_core/include” --include-path “/code/generative_models/style_gan2_play/venv/lib/python3.6/site-packages/tensorflow_core/include/external/protobuf_archive/src” --include-path “/code/generative_models/style_gan2_play/venv/lib/python3.6/site-packages/tensorflow_core/include/external/com_google_absl” --include-path “/code/generative_models/style_gan2_play/venv/lib/python3.6/site-packages/tensorflow_core/include/external/eigen_archive” 2>&1 “/data/tools/style_gan/stylegan2/dnnlib/tflib/ops/fused_bias_act.cu” --shared -o “/tmp/tmp3guzoekd/fused_bias_act_tmp.so” --keep --keep-dir “/tmp/tmp3guzoekd”

nvcc fatal : Value ‘sm_75’ is not defined for option ‘gpu-architecture’

I have cudnn 7.6.5 installed.

Here’s the nvidia-smi output:
±----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce RTX 208… Off | 00000000:42:00.0 On | N/A |
| 41% 38C P8 20W / 260W | 457MiB / 11016MiB | 2% Default |
±------------------------------±---------------------±---------------------+

Could it be a problem that the graphics drivers are 440, CUDA 10.2?

Thanks

Sorry for the false alarm.
I point /usr/bin/nvcc at the correct version of the nvcc binary (/usr/local/cuda/bin/nvcc) at all is well.