To know that new code produces the expected results

Just as with any research, we want to ensure our results are reliable, reproducible and trustworthy.

As software grows, our codebase becomes bigger and more complex.

When we rewrite and update code, we want to know that our new code works as intended, both in its basic functionality and as part of a bigger web of dependencies.

Writing tests for your software will save you time and headaches in the long run

Tests should be written alongside code, one test per ‘unit’ of code

In Python, a ‘unit’ would generally be considered to be a function or class

Any time that you manually run something and check the result

But! Tests can be written before we write any code, this is known as Test Driven Development (TDD)

By writing tests alongside your code, we know that as long as our tests are passing, our previously written code doesn’t need to be revisited