Focusing on Data Science & Less on Engineering and Dependencies
The Real Python Podcast
Structuring Python Projects and Python Packaging
The chapter discusses the benefits of using PyProject Toml file and PIP to set up Python projects. It covers topics such as structuring files and folders, running scripts, exploring the import system, and understanding the Python packaging world.
How do you manage the dependencies of a large-scale data science project? How do you migrate that project from a laptop to cloud infrastructure or utilize GPUs and multiple instances in parallel? This week on the show, Savin Goyal returns to discuss the updates to the open-source framework Metaflow.
Savin briefly describes the Metaflow platform and the goal of simplifying engineering overhead for data scientists and programmers. We discuss how the platform captures snapshots of a project as you work, allowing you to go back in time or share the state of your project with another team member.
We dig into the complicated process of managing dependencies for machine learning and data science projects. Savin describes how the required external libraries can be specified within a flow with the new @pypi or @conda decorators. This allows a project to scale from a local machine to the cloud or multiple instances with all dependencies included.
He talks about starting a new company, Outerbounds, with fellow co-workers from Netflix. Their vision is to continue to build the Metaflow open-source platform and offer customers scalable enterprise-grade infrastructure.
This week’s episode is brought to you by Intel.
Course Spotlight: Everyday Project Packaging With pyproject.toml
In this Code Conversation video course, you’ll learn how to package your everyday projects with pyproject.toml. Playing on the same team as the import system means you can call your project from anywhere, ensure consistent imports, and have one file that’ll work for many build systems.
Topics:
- 00:00:00 – Introduction
- 00:02:25 – Update on Metaflow
- 00:04:13 – What is Outerbounds?
- 00:07:26 – An ML platform to serve data scientists needs
- 00:13:02 – Dependency reproducibility via
@condaand@pypidecorators - 00:26:18 – Sponsor: Intel
- 00:27:10 – Storing lock files along with snapshots
- 00:29:17 – Working alongside code and dependency management systems
- 00:34:03 – Scaling a project from laptop to the cloud
- 00:40:13 – Video Course Spotlight
- 00:41:41 – Getting visibility on processes
- 00:47:23 – Adjusting your project due to GPU availability
- 00:52:27 – Example of jumping back into a project one year later
- 00:55:54 – What are you excited about in the world of Python?
- 00:57:39 – What do you want to learn next?
- 00:59:35 – How can people follow your work online?
- 01:00:19 – Thanks and goodbye
Show Links:
- Metaflow - a framework for real-life ML, AI, and data science
- Infrastructure for ML, AI, and Data Science - Outerbounds
- Human-Friendly, Production-Ready Data Science with Metaflow- Savin Goyal | SciPy 2022 - YouTube
- Episode #61: Scaling Data Science and Machine Learning Infrastructure Like Netflix – The Real Python Podcast
- New in Metaflow: The Long-Awaited
@pypiDecorator - Outerbounds - Managing Dependencies - Metaflow Docs
- Secure ML with Secure Software Dependencies - Outerbounds
- Directed acyclic graph (DAG) - Wikipedia article
- Visualizing Results - Metaflow Docs
- Seamless Data and ML Pipelines with Airflow and Metaflow - Outerbounds
- Episode #142: Orchestrating Large and Small Projects With Apache Airflow – The Real Python Podcast
- Savin (@SavinGoyal) - X
- Savin Goyal - LinkedIn
- Building the ML-driven future - Outerbounds Blog
Level up your Python skills with our expert-led courses:
The AI-powered Podcast Player
