Python Tutorial: DRY and "Do One Thing"

Want to learn more? Take the full course at https://learn.datacamp.com/courses/writing-functions-in-python at your own pace. More than a video, you'll learn hands-on coding & quickly apply skills to your daily work. --- DRY (also known as "don't repeat yourself") and the "Do One Thing" principle are good ways to ensure that your functions are well designed and easy to test. Let's see how. When you are writing code to look for answers to a research question, it is totally normal to copy and paste a bit of code, tweak it slightly, and re-run it. However, this kind of repeated code can lead to real problems. In this code snippet, I load my train, validation, and test data, and plot the first two principal components of each data set. I wrote the code for the train data set, then copied it and pasted it into the next two blocks, updating the paths and the variable names. But one of the problems with copying and pasting is that it is easy to accidentally introduce errors that are hard to spot. If you'll notice in the last block, I accidentally took the principal components of the train data instead of the test data. Yikes! Another problem with repeated code is that if you want to change something, you often have to do it in multiple places. For instance, if we realized that our CSVs used the column name "label" instead of "labels", we would have to change our code in six places. Repeated code like this is a good sign that you should write a function. So let's do that. Wrapping the repeated logic in a function and then calling that function several times makes it much easier to avoid the kind of errors introduced by copying and pasting. And if you ever need to change the column "label" back to "labels", or you want to swap out PCA for some other dimensionality reduction technique, you only have to do it in one or two places. However, there is still a big problem with this function. First, it loads the data. Then it plots the data. And then it returns the loaded