The last submission No 9715578 reported following error:
```
RuntimeError: 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
```
It works on my test system with RTX-3070, inside it uses following conda modules:
```
cudatoolkit 11.1.74 h6bb024c_0 nvidia
pytorch 1.9.0 py3.8_cuda11.1_cudnn8.0.5_0 pytorch
```
Created by Vladimir Fonov vfonov I already altered my submission to use CPU for inference.
My contained did run using GPU for inference on my local system. Hi @vfonov ,
I unfortunately do not know why you are receiving this error, as we do run the Docker submissions with the GPU attached, e.g.
```sh
docker run \
--rm \
--network=none \
--runtime="nvidia" \
-v /path/to/input:/input:ro \
-v /path/to/output:/output:rw \
```
When you run your container locally, does it run? What is the Docker run command you use?
I'm not sure if this will help as well, but the driver and CUDA versions used for evaluation are 470.57.02 and 11.4, respectively.
Drop files to upload
Docker container doesn't see NVIDIA driver page is loading…