Issue with CUDA

[kylepott]

Hi I'm getting an issue installing Cuda. Any suggestions?

~/3d-photo-inpainting-master: conda install pytorch==1.4.0 torchvision==0.5.0 cudatoolkit==10.1.243 -c pytorch
Fetching package metadata ...........

PackageNotFoundError: Package missing in current linux-64 channels:

• cudatoolkit ==10.1.243

[kylepott]

I was able to install the toolkit from the nvidia website, but now I get the following error.

RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

[kylepott]

I updated main.py to use torch.device('cpu') and now I get this error.

Found no NVIDIA driver on your system. Please check that you
have an NVIDIA GPU and installed a driver from
http://www.nvidia.com/Download/index.aspx

I'm running Ubuntu on an old MacBook Pro (~2011). I'm guessing I may have more luck spinning up an EC2 instance on AWS and trying again.

[kylepott]

I was able to install an NVIDIA driver, now error is:

AssertionError:
The NVIDIA driver on your system is too old (found version 9010).
Please update your GPU driver by downloading and installing a new
version from the URL: http://www.nvidia.com/Download/index.aspx
Alternatively, go to: https://pytorch.org to install
a PyTorch version that has been compiled with your version
of the CUDA driver.

[kylepott]

I now tried a brand new instance from AWS. I used the Ubuntu Deep Learning AMI (Free Tier) and the install went well except I had to update main.py to again use 'cpu' instead of device. Then when I ran my first image I got the following error:

RuntimeError: cuda runtime error (100) : no CUDA-capable device is detected at /opt/conda/conda-bld/pytorch_1579022060824/work/aten/src/THC/THCGeneral.cpp:50

[kylepott]

I updated the arguments.yml file to -1 for GPUs to just use CPU and now it appears to be running and then just says "Killed" with no other messages? ideas?

python main.py --config argument.yml
0%| | 0/2 [00:00<?, ?it/s]Current Source ==> 594A2731-768x512
initialize
device: cpu
start processing
processing image/594A2731-768x512.jpg (1/1)
torch.Size([1, 3, 256, 384])
finished
Start Running 3D_Photo ...
Killed

[kylepott]

So after a day of working on it, I realized I was just using a drastically under-powered and under GPU'd computers. I got everything working and I think it's truly amazing. Thanks for your work on this project. Marking my issue as closed.