Paper: Improving Code Quality with Array and DataFrame Type Hints

[flexatone]

If you are creating this PR in order to submit a draft of your paper, please name your PR with Paper: <title>. An editor will then add a paper label and GitHub Actions will be run to check and build your paper.

See the project readme for more information.

Editor: Charles Lindsey @cdlindsey

Reviewers:

• Victoria Adesoba @vicadesoba
• Talha Irfan @EngineerKhan

[github-actions]

Curvenote Preview

Directory	Preview	Checks	Updated (UTC)
papers/christopher_ariza	🔍 Inspect	✅ 24 checks passed (8 optional)	Aug 7, 2024, 3:27 PM

[EngineerKhan]

I have read this paper thrice and cannot find any issue with it. It's well-written, so easy to follow and builds a narrative in an excellent manner.

Some Minor Suggestions (can be ignored):

• "Just like a dictionary, a DataFrame is a complex data structure composed of many component types: the type of the index labels, the type of the column labels, and the types of column values." - the type of part is a bit redundant and can be rewritten.
• Raised errors are there in the code comments, but some of them (as per the author's discretion) may be explained a bit too.
• "The same annotations can be used for runtime validation. While reliance on duck-typing over runtime validation is common in Python, runtime validation is often needed with complex data structures such as arrays and DataFrames." - maybe we can give an example to support the claim?
• At some point, the section under "DataFrame Type Annotations" becomes a bit hard to read. Maybe we can add some subheadings there? Again, this is something the author can better determine.

Overall, a well-written article. Not only, I have learnt new things, but also enjoyed reading it with its smooth style.

[flexatone]

Many thanks, @EngineerKhan , for your suggestions and feedback.

[vicadesoba]

@flexatone Glad to review your paper and nice to meet you! Just a quick intro: I got introduced to the open source community during my time at makepath a geospatial data science company that maintains several open source packages and I'm involved with Project Bokeh (as much as time allows). Recently, I started working for a process engineering company that isn't involved in the open source community so volunteering with SciPy gives me a small way to stay involved!

[cdlindsey]

Thanks for the interesting paper. I think it's worth mentioning how these issues are handled (or not) with TensorFlow and Pandas.

[flexatone]

Many thanks, @cdlindsey , for your feedback. I have added a small section describing how these issues are handled in PyTorch, TensorFlow, and Pandas.