Python Data Analytics: From Notebooks to Production
59mAdvanced2024-08-26
Authors
Miki Tebeka
CEO at 353Solutions
Course details
Your code needs to run in production, which has its own set of challenges and restrictions. In this course, instructor Miki Tebeka shows you how to convert your research notebook to production ready code. Learn about the needs of production, where notebooks can really help you, and where they might fall short. Explore why you should organize your code and how to write clean API for your modules, as well as how to structure your code with sub modules. Find out why testing your code is important and go over useful techniques and best practices. Plus, dive into dependency management, logging and metrics, performance tuning, securing your code, and more.
Skills covered
Data Science FoundationsPythonData AnalysisProgramming LanguagesData ScienceBusiness Analysis and StrategyBusiness Software and ToolsOpen SourceSoftware DevelopmentOne-Off
Concepts
Introduction
- Notebooks to production
- What you should know
- Using GitHub Codespaces
Notebooks to Production
- Understanding production
- Where notebooks excel
- Where notebooks come short
Organizing Your Code
- Why you should organize your code
- Module API
- Sub modules
- Main
- Challenge - Convert logs notebook
- Solution - Convert logs notebook
Testing Your Code
- Why testing is important
- Running notebooks
- Parametrized tests
- Test fixtures
- Continuous integration
- Challenge - Test tags
- Solution - Test tags
Dependency Management
- The problems with dependencies
- Specifying and installing dependencies
- Separating test dependencies
- Distributing your package
- Challenge - Create an environment
- Solution - Create an environment
Running in Production
- Logging and metrics
- Configuration
- Performance tuning
- Securing your code
- Challenge - Make the monthly report production ready
- Solution - Make the monthly report production ready
Conclusion
- What's next