Setting up GPU for fastai_v3

@muhajir the steps listed will work if the nvidia drivers version is 396.xx+. So need to update nvidia drivers on top of the steps listed

thanks for listing down all steps, i followed all but no success, currently i am using cuda 8.0 version

Untitled2

I have gone through the steps, i am not able to open 1st link that you have mentioned,
i am using driver version: 387.34 and cuda 8.0

I would un-install every thing, install newest nvidia drivers, then install everything according to guide.

pytorch comes packaged with cuda by default so system cuda need not be changed…try updating the nvidia drivers to 396.xx+ for the OS ur using(…that will fix the issue…

1 Like

Which version of pytorch comes with cuda 10? I’d really like to leverage tensor cores and fp16. Thanks.

1 Like

Hi Raghav,
Sorry for late post,

I have doubt about version compatibility of cuda 8.0 with nvidia driver version.
Is cuda 8.0 will support nvidia driver version 396.xx+?
or
do i need to install cuda 9.0 for nvidia drivers version 396.xx+?

currently i am following this post:

steps:
till now i have uninstall nvidia-driver 387.34 on cuda 8.0. and now the next step is to install 396.18 on cuda 8.0.
os used: ubuntu 16.04

@shwetap7 forget about cuda for now, just upgrade nvidia drivers to 396+…pytorch comes bundled with cuda inherently unlike tensorflow so fastai as it sits on top of pytorch should work

thanks will try that…

Hi, could you tell me exact 396 version you are using? I mean what is xx in 396.xx?

@ymittal23 the exact version im currently using is 396.54

Thanks for help.

I am getting Floating point exception (core dumped)
on running this in lesson 1:

interp = ClassificationInterpretation.from_learner(learn)

I tried it with cuda 9.0 and cudnn 9.1
also with cuda 9.2 and cudnn 7.4
Didn’t got the issue

Got the same error in all configs

Check this thread

I believe this one from @cpbotha tutorial

1 Like

Hi,
I can’t make the code of lesson1-pets run on my GPU, it is defaulting back and using the CPU.

Using the same conda environment with keras and tensorflow it works with the command
os.environ[“CUDA_VISIBLE_DEVICES”] = ‘0’
But this command seems not to be working with fast.ai…

Another attempt was:
device = torch.device(‘cuda’ if torch.cuda.is_available() else ‘cpu’)
print(Using device: device)
This is printing: Using device cuda
But the CPU gets really loud so I can tell that the model is trained on the cpu.

Any suggestions?

1 Like

Hi, I just started and have the same issue. I was wondering if fastai was using my GPU or not. I end up using the util gpu-z where you can have a look at gpu clock and gpu load and then I can see that as the model train the gpu clock increase to handle the work load.

So try that to see if your is using gpu or not.

Cheers

Try checking GPU usage with nvidia-smi command to make sure if GPU is being used or not

The GPU seems to be used by my environment.
The process:
0 9492 C …n\Anaconda3\envs\dl-gpu_py36\python.exe N/A
is listed.

I tried all the above steps twice:
in the first try i was unable to do it because of some connectivity issue
but the second time when i started and along with the git clone command I added a file name “deeplearning” as it was showing me that there already exists a file there and hence path destination error of some kind
but now i am not able to find that whole folder anywhere?
what should i do now?
I’ve searched everywhere…
plzzz hellpppppp