Skip to content

Latest commit

 

History

History
45 lines (31 loc) · 2.61 KB

dev_environment.md

File metadata and controls

45 lines (31 loc) · 2.61 KB
Raw

Development environment on Docker

To support development without focussing too much on configuration, we have set up a Docker Compose environment that can be set up with 1 command.

When using your GPU, there's no need for CUDA or CUDNN to be installed. OS-level NVIDIA drivers are CUDAnough good enough as CUDA and CUDNN is inside the container.

Whilst not required, this setup gives you the exact same environment and dependencies as everyone else. This will avoid the "works on my machine" type of bugs.

How to get started

Todo: The author has a Linux-first setup. If you use Windows, please help to update this readme to work for Windows, too.

Requirements:

Installation:

  • Clone this repo:

    • git clone https://github.com/LAION-AI/medical.git
    • or git clone [email protected]:LAION-AI/medical.git if using SSH.

  • cd into the cloned repo.

  • Run docker-compose up -d. It should build the image and spawn the container. This might take a while depending on your internet connection.

  • Now, run docker-compose logs --tail=100 to get the logs in your terminal. Inside, you will find a URL and a token:

    laion_medical_research_1  |     To access the server, open this file in a browser:
laion_medical_research_1  |         file:///home/dev/.local/share/jupyter/runtime/jpserver-1-open.html
laion_medical_research_1  |     Or copy and paste one of these URLs:
laion_medical_research_1  |         http://hostname:8888/lab?token=a867047cb0418730f08967096a80b2c6f8acb624e01204d2
laion_medical_research_1  |      or http://127.0.0.1:8888/lab?token=a867047cb0418730f08967096a80b2c6f8acb624e01204d2

  • Copy and paste the link pointing to http://127.0.0.1:8888 in your browser. You should now have access to all the notebooks with local inference/training.

Keeping dependencies up to date

If you update dependencies (for example, because we need a new dependency), the best way to do it is like this:

  • Open the terminal inside the container (either in JupyterLab or docker-compose exec laion_medical_research bash)
  • Run pip install <package name> to install your dependency
  • Run pip list --format=freeze > Docker/requirements.txt to update the dependencies. Now you can commit and push this to the repo.
    • Don't use pip freeze! This is because of an issue with pip corrupting the requirements.txt at the moment.
  • Others can run docker-compose up -d --build to update their containers with the new dependencies.