Design Patterns for Data Science, Part 2: Paper vs. Ceramic

In the last Design Patterns for Data Science post, we talked about the relative advantages of Jupyter notebooks and scripts, and we moved some code from a preprocessing notebook to a method inside a python file. When we did that, we discovered that the limitations we faced changed from tooling limitations to programming limitations—problems we can solve with design patterns from software engineering.

But before that, I want to talk for a moment about data science, programming, and cups.

I had dinner with a friend the other night, a principal data scientist with a software engineering background. He shared some valuable insight on the programming philosophies of software engineers versus data scientists. Data scientists, he pointed out, tend to build code to have a short lifespan: a script to change the data once, or a helper method to use for few weeks until a model is finished. Software engineers, by contrast, imagine their code having a longer lifespan: months or, if it’s really built well, years. Many old-hand software engineers have a story about some embarrassing code they wrote that accidentally stayed in use for decades. Decades. 

So it’s not that one group is better at writing code than the other. It’s that one group has optimized on programming paper cups, and the other group has optimized on programming ceramic mugs.

paper coffee cup and ceramic mug

But there’s a lesson in the stories from the old-hand software engineers: code often sticks around far longer than anybody thought it would. That’s true in software engineering and has resulted in a number of patch job solutions from the cute ones (like Android’s nullColumnHack) to the scary ones (like security holes in the web itself).

It’s also true in data science: a temporary solution ends up getting passed around and reused without context. The data science workspace ends up littered with paper cups—dented, coffee-stained, oft-reused paper cups, each of them well past their intended lifespan.

Data science can benefit from learning some of the techniques that software engineers use to program mugs. Engineering techniques help us better prepare our code for the reuse, repurposing, and context sharing that it can expect to withstand.

That’s why I’d like to share the object structure I ultimately came up with for solving the case study from the last post.

Review from the Last Post

Here’s our case study:

Suppose we have 50,000 rows of labeled data. Each row has a paragraph of legal text in a column called base_content. We want to use this base_content to classify the rows into one of twelve categories of specific types of legal text. We need to extract the data out of large CSV files, and then we need to do a little preprocessing to get them ready for testing models.

That post walked through several options:

  1. Doing the preprocessing at the top of the analysis notebook
  2. Importing a notebook into another notebook (please don’t)
  3. Saving preprocessing steps to a csv
  4. Copying preprocessing code into a method in a python file

The last solution has a number of advantages, and that’s where we move from tooling limitations to programming limitations:

1. It starts the extraction process from scratch every time we run it. We’re dealing with 48,000 rows of text, each of which has a text attribute that needs tf-idf vectorization. The method you see above takes roughly 20 seconds to run. That feedback loop is long enough to add friction to our workflow. And every time we need to undo our notebook changes and get a fresh copy of the dataframe, we have to wait that 20 seconds again. The ideal solution would be much faster than this.  

2. We do not get to make any choices about how our preprocessing is done for each analysis step. Back at Option 1, we talked about how these data analysis projects go. Each analytical step requires some unique modifications to the data. It’s possible that we want to start with different pre-processing steps depending on the analysis we’re doing. With this method, though, we’re always getting the same steps—including tf-idf, which is computationally expensive. Ideally, inside a client notebook, we could compose together the pre-processing steps we need without any additional ones we don’t need.

So what did I end up doing with this data access and pre-processing code?

Option 5. Singleton data storage wrapped in a class with a fluent interface

Singleton: The data exists exactly once in a shared data store. Class: A distinct instance of the data—but all of the instances reach into the single shared data store. Fluent interface: Each method on an instance returns the instance itself so you can chain methods together. You have used a fluent interface before if you have chained pandas methods together like this: df = pd.read_csv('some.csv').rename(columns=str.lower).drop('unnamed: 36', axis=1).pipe(custom_method)

Let’s take a look at the code itself to get an idea of what is happening. I have added comments within the individual methods in case you’re interested in a deep dive:

As you can see, when we instantiate a Content(), we refresh a data cache. _cached_data lives on the Content class itself, rather than any given instance, so once one instance has fetched it, all instances have access to it. For this reason, the first time one of our notebooks makes a Content(), it takes some time to fetch the data. The second time and every time thereafter, though, the data is already loaded for use! Check out these two time trials:

The first time takes about 20 seconds. The second time? 0.0004 seconds.

Singletons introduce a risk: if one instance modifies the store, then every instance now has a modified store—even if they did not want the data modified. Our class here mitigates that risk because all methods that modify the underlying data are both idempotent and agglutinative.

Idempotent: We can run it 1 time or 400 times, and the result will be the same.

Agglutinative: All changes add to the data store. No methods change or remove data in the data store. Data doesn’t get re-represented in the notebook no matter what order a notebook user chooses to run their cells. Furthermore, we know that tf-idf takes a long time. So if we don’t need that pre-processing step, we run Content().to_dataframe() and we don’t have to wait for it to finish at all—not even the first time we access the data. We can take advantage of python’s built-in testing tools to write tests for this class and ensure that we don’t cause regressions (break things that were previously working) if we make changes to the existing code.


Software engineers spend a lot of time thinking about how to make code flexible and reusable. The data science community can benefit from that thought. Here we have taken a method inside a python class—which gets the job done, albeit inefficiently—and converted it into a python class with a single instance of expensive-to-access data. We have also added a fluent interface so that clients can customize which pre-processing steps they run on the data. In future posts, hopefully we’ll get to see additional examples of ways that the software engineering and data science words can cross-pollinate to make everybody happier.

If you liked this post, you might also like:

This post explaining cost function optimization. People seem to like it 🙂

This post explaining Taylor Series. Someone once told me that this post ‘saved their life,’ which was probably an exaggeration but I WILL TAKE IT, THANK YOU.

Visualizing the regression and classification process (for presenting models to businessfolk/helping SMEs with error analysis)


  1. Hi,
    Could you explain the correspondence of Option 5 with solving this: “And every time we need to undo our notebook changes and get a fresh copy of the dataframe, …”? I think I’m misunderstanding something here but wouldn’t needing a fresh copy of the dataframe mean we have updated df (fetched from to_dataframe() ). And if so, wouldn’t the cached copy be updated too?

    • Hi Bijay! Thanks for your comment!

      Indeed, when we call .to_dataframe() we’re getting the cached data (which happens to be stored as a dataframe) from the Content() object. But until we call .to_dataframe(), the methods we’re calling in that chain are called on the Content() object.

      When we say df = Content().to_dataframe() (with or without methods in the middle), we’re getting a copy of the data cached in Content()—not the cached instance itself. So when we call dataframe methods on that df, we’re modifying our copy.

      The way that we modify the cached data itself is to call methods defined on Content() that refer to the cached instance inside the class. In our example, the only modifications the client can make are the ones that run the first time we instantiate a Content() and the first time we call .tfidf_encoded(). You could modify the cached data directly by doing Content._cached_data = df.some_modification(), but that’s not the intention. In python, we use a leading underscore on methods and attributes to indicate that this thing is meant to be private to its file.

      Otherwise, what we’re doing is saying “take your cached single instance of data and give me a copy of that.” And that’s a lot faster than re-making it each time we need it!

      • “When we say df = Content().to_dataframe() …, we’re getting a copy of the data cached in Content()”.

        I mean to say `to_dataframe()` needs to have “return Content._cached_data.copy()” for this to happen, right? Else it’s not a copy; df refers to exactly the same object as _cached_data and modifications on df are modifications on _cached_data.

      • Well butter my biscuit, you’re right! It looks like I published this post with an outdated version of the code in my gist. Thank you for catching that!

        The example has been updated. If you have a look, you’ll see your name attributed in there 🙂

      • Np. I’m trying to figure out/learn a good workflow around notebooks and these posts are really helpful.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.