Software Engineering for Data Scientists

Chapter 10. Sharing Your Code: Version Control, Dependencies, and Packaging

Sharing your code and collaborating with others is hugely important to your success in data science. You might join an existing project that has a large codebase. You might start by working on your code in isolation, but as your project grows, you may want to share it with others so that they don’t need to solve the same problems as well. Or you could be contributing to open source projects.

If you’re sharing your code publicly, as an open source project, that allows you to join a community. Python open source libraries are a huge ecosystem, and data science would be so much harder without pandas, NumPy, scikit-learn, Matplotlib, and many others.

Whether you’re contributing to open source projects or collaborating with teammates at your company, you’ll need to know some tools and techniques that make it easier for multiple people to work on the same code. The tools may vary, but many of the principles are standard.

In this chapter I’ll cover these tools and techniques that will help you collaborate with other developers. First, version control is important because it’s the standard way of collaborating on a codebase. Second, I’ll cover tools for managing the third-party libraries that your code depends on. Finally, I’ll describe how to package your projects so that other people can install them.

Version Control Using Git

Version control is a way of tracking what changes have been made to a codebase, and ...