docker.md

Docker

Instead of installing the body composition code directly on your system, you can use the Dockerfile in the repository to build a docker image that already has all the code and dependencies built in. If you are unfamiliar with docker, I suggest reading this guide and familiarizing yourself with docker before continuing.

Building the Docker Image

After you have cloned the body composition repository, change directory to the root of the cloned repository (where the file called Dockerfile is found) and run:

$ sudo docker build -t body_comp:latest .

The process of building the image will take several minutes to complete.

Running the Container

Once built, you should start an interactive session within a container in order to run the code. Remember that you will need to mount in the locations of any data you want to access from inside the container using the -v option, as well as anywhere you want to write the results to. Writing the results within the container's filesystem means they will be deleted when the container is shut down.

For example, the following command will run the container with data on your system (/path/to/data) mounted into the container so that it's visible at /data:

$ sudo docker run -it -v /path/to/data/:/data/ body_comp:latest bash

Running Code Inside the Container

After executing the command above, you should find yourself in the /bin/body_comp directory in the container. You will find all the files for running the model in this directory (as described in the other documentation pages). The environment should all be set up and you should be able to run these python files straight away.