Future Wisdom @ GSAIS (Shishu-Kan) , Kyoto U.

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
Last-updated: 2025/02/23 liangz

====== Docker tutorial ======

===== Step 1: Understand docker =====

  * Read https://docs.docker.com/get-started/docker-overview/ and understand (docker) image, container, daemon, client, hub (registry), pull, push.
  * Install docker (Linux Mint as example. Information on other OSes are welcome too.)
    * $ sudo apt install docker.io
    * Add your username to the "docker" group (or by editing /etc/group directly).
      * $ sudo usermod -aG docker your_username
      * Logout & login to make the change of group into effect, or simply reboot your Linux.
  * Run a ubuntu container (docker client will automatically pull a ubuntu image and create a local container.). If you get a console with command prompt "root@xxxxxxxxxxxx:/," it worked.
    * $ docker run -i -t ubuntu /bin/bash
  * Exit the container by exit (notice the prompt changed to "#", showing you are the root)
    * # exit

===== Step 2 (if you have an NVidia GPU): Install Nvidia GPU support =====

  * Follow https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

===== Step 3: Install PyTorch (for development) =====

  * $ docker pull pytorch/pytorch:latest
  * $ docker run --rm --gpus all pytorch/pytorch:latest python -c "import torch; print(torch.cuda.is_available())"
  * If you see "True", it worked.

===== Step 4: Work with PyTorch =====

  * $ docker run -it --gpus all --rm pytorch/pytorch:latest bash
  * # nvidia-smi
  * If you see the use of your GPU, it worked.

===== References =====

  * https://github.com/saikhu/Docker-Guide-for-AI-Model-Development-and-Deployment (Chapter 6 is obsolete. See the above Step 2 for the latest information)