Things You Can Do with Python: Advanced and Special Use Cases

Python is instrumental in so many data science and machine learning workflows that it can sometimes just blend into our daily rhythm; how often, after all, do you think about your office light switch or door knob? You use them all the time, too.

For our first Python-centric Variable edition of 2024, we decided to focus on some of the more interesting and off-the-beaten-path use cases we’ve published recently. We love a good Pandas or Matplotlib tutorial—and so do many of our readers—but sometimes it’s fun to take a break from bread-and-butter topics and dive into some fancier stuff. This week, let’s indulge a little! We hope you enjoy the nine Python reads we’ve selected, which cover a striking range of projects and challenges.

• How to Build a Graph-Based Neural Network for Anomaly Detection in 6 Steps
  For her latest step-by-step tutorial, Claudia Ng, who often writes about interesting Python-based projects, focuses on building a graph-based neural network that can handle heterogeneous graph data for link prediction.
• Introducing the Quad-Tile Chart & Squaremap: Squarify Your Data
  Why settle for existing visualization formats when you can create your own? Nick Gerend invites us on a behind-the-scenes tour of the process behind building Quad-Tile Chart, his Python-powered, axis-free approach for visualizing a set of values as squares.
• Tagging Mountaineering Accident Reports Using bart-large-mnli
  Still on the innovative-visualization front, Karla Hernández walks us through the steps of putting together a stunning data-storytelling artifact that highlights findings from a custom-tagged mountaineering-accident dataset.
• Finite Automata Simulation for Leveraging AI-Assisted Systems
  Using finite-state machines, Sofya Lipnitskaya explores a potential approach for optimizing the performance of complex real-world AI-assisted processes — in this case, an object-detection system that would activate water sprinklers to scare away invading chickens!

• Watching Storms from Space: A Python Script for Creating an Amazing View
  Working with geospatial data comes with its own set of challenges; Mahyar Aboutalebi, Ph.D.’s latest guide unpacks the process of building a Python script that allows you to collect satellite images and transform them into powerful storm animations.
• Python’s Most Powerful Decorator
  In case you missed it, Siavash Yasini’s detailed introduction to Python’s @property decorator is one of our most-read programming articles in recent weeks. It covers several useful ways to leverage its power: from protecting data attributes from being overwritten to lazy-loading and memory optimization.
• Molding the Imagination: Using AI to Create New 3D-Printable Objects
  After text, image, music, and video, could 3D objects become the next frontier for generative AI? Robert A. Gonsalves shares the results of his recent experiments, which depend on Midjourney for image generation and on some good-old Python code for translating these into tangible objects.
• Text Embeddings: Comprehensive Guide
  If you’re new to the world of text embeddings, Mariya Mansurova’s primer is a great place to start—it’s both (very) thorough and accessible, and the hands-on sections include all the Python snippets you’ll need to start tinkering on your own.
• Understanding Junctions (Chains, Forks, and Colliders) and the Role they Play in Causal Inference
  In his recent deep dive on DAGs (directed acyclic graphs), Graham Harrison zooms in on junction types and their importance in causal-inference tasks. Along the way, he also demonstrates how to generate datasets, execute ordinary least squares (OLS) regression, and more, all with—you guessed it—Python.

Thank you for supporting the work of our authors! If you’re feeling inspired to join their ranks, why not write your first post? We’d love to read it.

Until the next Variable,

TDS Team

Things You Can Do with Python: Advanced and Special Use Cases was originally published in Towards Data Science on Medium, where people are continuing the conversation by highlighting and responding to this story.