Adventures in Machine Learning

Moving from Dev Notebooks to Production Code - ML 098

Dec 22, 2022
Guest Mike Arov discusses challenges transitioning ML projects from notebooks to production code, emphasizing tools like Linear Pi. Topics include bridging ML Ops and DevOps, managing software versions, automating artifact generation, and tracing dependencies in notebooks. The conversation highlights the importance of collaboration in open-source projects and the significance of automation tools in streamlining deployment processes.
Ask episode
Chapters
Transcript
Episode notes
1
Introduction
00:00 • 4min
2
Bridging the Gap Between ML Ops and DevOps
03:33 • 18min
3
From Notebook to Production: Challenges and Evolution
21:13 • 9min
4
Automating Artifact Generation for Productionized Code with Lineage
30:24 • 2min
5
Managing Software Versions and Serialization Challenges in Python
32:43 • 9min
6
Discussion on Managing Dependencies and Tracing Dependencies in Notebooks
41:30 • 2min
7
Transitioning from Development Notebooks to Production Code in Machine Learning Projects
43:32 • 25min
8
Freelancing Opportunities, Importance of Tests, and Lenny and Pi Project Discussion
01:08:16 • 5min