But It Works on My Machine! How to Make Our Analyses...
summary
Guidance for making Python data-analysis projects reproducible by focusing on clear communication of components and where to find necessary pieces. Covers practical practices: using virtual environments, version control, directory/file organization, writing reusable functions and automation, and creating useful documentation, aimed at researchers and data scientists (including beginners).
AI-written summary · about
chapters (17)
- 0:00Welcome
- 1:53Speaker intro
- 2:41Reproducibility issues
- 7:55Reproducibility definition
- 9:53Environment control
- 11:15Version control
- 12:42Directory structure
- 14:38File and folder names
- 17:26Alphabetical order
- 18:15Naming patterns
- 20:32Descriptive figure names
- 21:58Data testing automation
- 23:46Clear documentation
- 25:25Context adaptation
- 26:21Recommended resources
- 27:40Final thoughts
- 28:10Q&A
related talks
41mBetter Code for Data Science
56mML Pipeline: Structuring Data Analysis Projects
30mFrom Chaos to Confidence: Solving Python's Environment Reprodu...
01h 14mUp your Bus Number - A Reproducible Data Science Workflow
38mIt Works on My Machine: Writing Python Code for Any Environment
45mSolving problems by sharing them... with Python!
38mBuilding Data Pipelines in Python
27mNotebook To Production