Install OpenDevin with Docker

Question

I am having difficulties understanding how to install OpenDevin with the provided docker image. In the README on GitHub, the following first installation steps are mentioned:

Installation

First, make sure Docker is running:

docker ps # this should exit successfully

Then pull our latest image here.

docker pull ghcr.io/opendevin/sandbox:v0.1

Then start the backend:

export OPENAI_API_KEY="..."
export WORKSPACE_DIR="/path/to/your/project"
python -m pip install -r requirements.txt
uvicorn opendevin.server.listen:app --port 3000

I do not understand where I "start the backend"? Do I need to run something like docker run (...) in a terminal window and then type these commands? Or do I have to clone the repository as well in order to have the requirements.txt file available for Python to know what modules it needs to install?

If I have to clone the repo locally, what's the goal then in pulling a Docker image? Isn't the goal of Docker to have everything in a "container", so that everything is in one place? Wouldn't the Python module have to be already installed in that Docker image?

I have Docker, Python and Node.js installed.

Answer 1

I got it to work thanks to this video:

[OpenDevin: AI Software Engineer With Complex Coding Completion][1]

One first needs to clone the OpenDevin repo and run requirements.txt from within this repo.