You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Sep 24, 2020. It is now read-only.
This repository was archived by the owner on Sep 24, 2020. It is now read-only.
I have the exact same config as mentioned by you in the readme but I get this error - tensorflow-1.8.0-cp36-cp36m-linux_x86_64(cuda9.2+cudnn7.1.4).whl is not a supported wheel on this platform. #2
I have cuda 9.0 installed too and I have /usr/local/cuda pointed to cuda-9.0 dir. But in my bashrc, I updated env params as following -
I have ubuntu 18.04 with Nvidia driver version 390.48
cuda 9.2, cudnn 7.1, nccl 2.2.13.
python 3.6.5
Also, did you also compile the binary with nccl support?
Thanks.