The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) DevOps for ML with Dotscience - #320
Nov 26, 2019
Luke Marsden, Founder and CEO of Dotscience, shares insights on streamlining MLOps for machine learning projects. He discusses the integration of MLOps and DevOps, highlighting the challenges faced in collaboration and reproducibility. The conversation dives into a manifesto that promotes software engineering practices in ML, aiming for better accountability and continuous deployment. Luke also explores features enhancing collaborative workflows and the benefits of using Jupyter for data science, along with containerized deployment strategies using Docker for optimized model performance.
AI Snips
Chapters
Transcript
Episode notes
ML Complexity
- Machine learning models involve data, models, and training software, making them complex.
- Tracking datasets, parameters, and metrics is crucial for managing this complexity.
MLOps Manifesto
- DotScience's manifesto emphasizes reproducibility, accountability, collaboration, and continuous delivery in ML.
- These principles are key for maturing ML processes to the level of software engineering.
Enterprise ML Challenges
- Many enterprises struggle to deploy ML models, highlighting the importance of continuous delivery.
- Reproducibility and accountability become critical as models are deployed and teams grow.