New data science tools and state-of-the-art models make a splash on a daily basis, but Python, for all its frequently cited shortcomings (speed, anyone?), remains one of the main unifiers of data and ML practitioners around the world. Scroll down the TDS homepage on any given day, and you’ll find it impossible to contest the programming language’s ubiquity and versatility.
To help you discover libraries, use cases, and optimization techniques you might not be familiar with yet, we’ve gathered a selection of recent Python-focused tutorials we found particularly well executed. They cover a lot of ground, so regardless of your role or background, you’re likely to find a new topic or workflow to explore. Happy learning!
- As a member of the pandas core team, Patrick Hoefler often hears about the popular library’s pain points; his new tutorial walks us through the nitty-gritty of using PyArrow (which pandas 2.0 supports) to address issues around working with non-standard and arbitrary dtypes. (As an added bonus, it also includes helpful pointers for Dask users.)
- If you’d like to expand your geospatial-data toolkit, don’t miss the latest post from Parvathy Krishnan (and coauthors Mahdi Fayazbakhsh and Kai Kaiser), which focuses on Digital Elevation Models (DEMs) and shows how to make the most of Python’s elevation package to explore and visualize them, as well as to calculate derivatives such as slope and elevation contours.
- Python provides us with many other ways to understand the world—or simply to see it more clearly. Case in point: Conor O’Sullivan‘s new guide for processing satellite images to remove those pesky clouds that sometimes (read: often) block our view.
- With healthcare systems still under considerable strain across the world, there’s a real need to optimize the allocation of staff and resources. Meagan Voulo presents a Python-powered analysis of social determinants of health (SDOH) as a potential way to forecast peak emergency-room usage.
- If you’d like to improve the performance of your code (and who wouldn’t?), Peng Qian is here to help with a concise tutorial for asyncio users: it presents best practices for approaching concurrent tasks in Python with the asyncio.gather, asyncio.as_completed, and asyncio.wait APIs.
- We promise we’re not trying to stir any pots, but… sometimes Python might not be the best option for your project, and that’s ok! Knowing when not to use Python is a key skill in its own right. Stephanie Lo‘s comprehensive guide will help loyal Pythonistas transition to R when needed, and details the differences you should be aware of when coding in another language.
Programming is great, but so are all these other posts we’d love for you to read this week:
- Our latest Monthly Edition is out, and if it has one goal, it’s to inspire you to design wonderful data science and ML projects.
- Molecular biology is one of the fields that has made the biggest strides with the aid of AI, and Serafim Batzoglou‘s masterful overview covers the past, present, and future of this interdisciplinary cross-pollination.
- The differences between reactive and proactive data teams might be more subtle than you think, but, as Barr Moses explains, there’s very little doubt as to which brings more value to the business.
- We’re now firmly in ChatGPT plugin season, with new ones popping up at a rapid clip. To get a handle on plugins’ benefits and risks, don’t miss Mary Newhauser‘s new deep dive.
- "Just as people forget, so do machine learning models – in particular, Large Language Models." In a powerful TDS debut, Matt Tengtrakool reflects on the role of forgetting—catastrophic forgetting, to be precise— in the context of machine learning.
Thank you for supporting our authors! If you enjoy the articles you read on TDS, consider becoming a Medium member – it unlocks our entire archive (and every other post on Medium, too).
Loyal TDS readers might be happy to know that our friends at Medium are planning a celebration of community and storytelling later this summer, and your input is encouraged to make it a successful event for everyone.
Until the next Variable,
TDS Editors
