Packaging Data Analyses & Using pandas GroupBy | Real Python Podcast #217

What are the best practices for organizing data analysis projects in Python? What are the advantages of a more package-centric approach to data science? Christopher Trudeau is back on the show this week, bringing another batch of PyCoder's Weekly articles and projects.

We discuss Joshua Cook's recent article "How I Use Python to Organize My Data Analyses." The article covers how his process for building data analysis projects has evolved and now incorporates modern Python packaging techniques.

Christopher shares his recent video course on grouping real-world data with pandas. The course offers a quick refresher before digging into how to use pandas GroupBy to manipulate, transform, and summarize data.

We also share several other articles and projects from the Python community, including a news roundup, working with JSON data in Python, running an Asyncio event loop in a separate thread, knowing the why behind a system's code, a retro game engine for Python, and a project for vendorizing packages from PyPI.

This episode is sponsored by Mailtrap.

Topics:

- 00:00:00 -- Introduction
- 00:02:18 -- Setuptools Breaks Things, Then Fixes Them
- 00:04:57 -- PEP 751: A File Format to List Python Dependencies
- 00:07:04 -- Python 3.13.0 Release Candidate 1 Released
- 00:07:15 -- Python Insider: Python 3.12.5 released
- 00:07:22 -- Django 5.1 released - Django Weblog
- 00:07:27 -- Django security releases issued: 5.0.8 and 4.2.15
- 00:07:49 -- How I Use Python to Organize My Data Analyses
- 00:13:45 -- Sponsor: Mailtrap
- 00:14:21 -- pandas GroupBy: Grouping Real World Data in Python
- 00:20:33 -- Working With JSON Data in Python
- 00:25:01 -- Asyncio Event Loop in Separate Thread
- 00:30:33 -- Video Course Spotlight
- 00:31:47 -- Habits of great software engineers
- 00:49:17 -- pyxel: A Retro Game Engine for Python
- 00:52:36 -- python-vendorize: Vendorize Packages From PyPI
- 00:54:18 -- Thanks and goodbye