I'm trying to use my GPU to train a network. I installed PyTorch and CUDA with anaconda. I have an NVIDA GTX 860M on my computer I would like to use.
When using the command :
$ lspci | grep NVIDIA
I get the output :
01:00.0 3D controller: NVIDIA Corporation GM107M [GeForce GTX 860M] (rev a2)
So the graphic card is there. But in a Jupyter notebook, the following lines :
print(torch.cuda.device_count())
print(torch.cuda.is_available())
print(torch.version.cuda)
Return :
0
False
9.0.176
I have tried updating the latest Cuda driver using conda it did not solve the issue. My PyTorch version is 0.4.1.post2. I couldn't find an answer browsing through the different online forums. I am on Ubuntu 18.04.1 LTS. Thanks' in advance !