Software Engineering for Data Scientists

Chapter 15. Next Steps

Congratulations on making it to the final chapter! I hope that after reading this book you’ll feel empowered to write great code that satisfies all the requirements you need and works well at large scale.

Next time you sit down to write some code, remember the five main points that I introduced in Chapter 1:

Simplicity

It’s easier for other people to use your code if you keep it simple.

Modularity

Break your code down into sensibly sized chunks. Use the principles of object-oriented programming that I described in Chapter 4 if they are a good fit for the problem you’re working on. Make sure the overall structure works well, as I described in Chapter 8.

Readability

Write documentation for your code, as I discussed in Chapter 9. Your code will be read much more often than it is written. Consistent formatting also helps with readability, as explained in Chapter 6.

Efficiency

Make sure your code is as efficient as it needs to be, using the methods in Chapters 2 and 3.

Robustness

Make sure that your code handles errors well, and that these errors are logged, as I described in Chapter 5. Testing, as I described in Chapter 7, is an extremely important skill to learn and put into practice.

These principles will help you move from one-off, ad-hoc, experimental code to reproducible, scalable code. Scalable code can be shared in packages, as I described in Chapter 10, or by deploying it as an API, as discussed in Chapters 11 and 12.

I’ve touched on a few ...