Best Practice for Data Science applications

General Discussions
1

I followed this article on Medium to set up my Docker environment (using Dockerfile, docker-compose.yaml, requirements.txt) to set up Tensorflow development environment: Tensorflow with GPU on Windows with WSL and Docker | by KNuggies | Medium

Here are my two initial questions:

  1. If I want to update the pip packages, do I just modify the requirements.txt file and then do “docker-compose up”?

  2. What is a good practice for setting up these 3 files (Dockerfile, docker-compose.yaml, and requirements.txt)? For example, do I put them for each project and so I create new ones under each project folder?

Thanks.
David