The GPU cannot be used in the container

[Olalaye]

I'am just following this link

• Current repo: run sudo docker pull ultralytics/yolov5:latest
• Common dataset: none
• Common environment: Docker image.

🐛 Bug

The GPU cannot be used in the container.

To Reproduce (REQUIRED)

Input:

# pull image 
sudo docker pull ultralytics/yolov5:latest

# run container
sudo docker run --ipc=host --gpus all -it ultralytics/yolov5:latest

# run command
python -c "import torch; print(torch.cuda.is_available())"

Output:

/opt/conda/lib/python3.8/site-packages/torch/cuda/__init__.py:52: UserWarning: CUDA initialization: CUDA unknown error - this may be due to an incorrectly set up environment, e.g. changing env variable CUDA_VISIBLE_DEVICES after program start. Setting the available devices to be zero. (Triggered internally at  ../c10/cuda/CUDAFunctions.cpp:104.)
  return torch._C._cuda_getDeviceCount() > 0
False

Expected behavior

Use the GPU in the container.

Environment

• OS: Ubuntu 20.04
• GPU: GeForce RTX 3090

Additional context

Add any other context about the problem here.

[github-actions]

👋 Hello @Olalaye, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@Olalaye this may be an indicator of underlying issues in the Docker install itself, or possibly the underlying CUDA install I think.

I verified the Docker image with a 3090 instance just now from vast.ai, and your test command, the nvidia-smi command, and a detect.py inference command all work well, so I'm not able to reproduce this problem.

vast.ai RTX 3090 instance rental

Note that all vast.ai instances are Docker images, with docker image name specified in upper left.

SSH commands inside Docker container

[glenn-jocher]

@Olalaye found this with a Google search:
pytorch/pytorch#47038

[Olalaye]

@glenn-jocher Thank you very much for your reply.
I avoided the problem by rebooting the computer, but I still don't know what went wrong.
Unfortunately, I need to do the same next time.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.