Torch.cuda.is_available() is False

#1

Hello,

I’ve installed Pytorch according to https://pytorch.org/get-started/locally/

by running

conda install pytorch torchvision cuda92 -c pytorch

and video card is installed

lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] (rev a1)

But still for some reason I get False

Python 3.6.6 |Anaconda, Inc.| (default, Jun 28 2018, 17:14:51) 
[GCC 7.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.cuda.is_available()
False
>>>

How to fix it?
Thank you!

#2

since 9.2 is fairly recent my hunch is that your driver is out of date. See here for tips: https://devtalk.nvidia.com/default/topic/1035692/ubuntu-16-04-cuda-9-2-with-driver-390-59-insufficient-driver-/