From POC to Prod, a data science project laid bare
summary
Covers a full data-science project pipeline from proof-of-concept to production, focused on deploying anomaly-detection models. Describes chosen machine-learning methods and practical industrialization topics: microservices architecture, running Python and R together in production, Kubernetes clusters, continuous integration, and monitoring.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 7:08Anomaly detection theory
- 13:53Modeling and interfaces
- 14:15Industrialization and API
- 20:58Automation and tests
- 27:16Deployment and integration
- 30:51Recommendations and conclusion
related talks
23mFrom PoC to Viable Product: industrializing a Big Data project
29mData Science Best Practices : From Proof of Concepts to Production
17mProductionizing Data Science Workloads
34mFrom POC to production: Strategies for achieving machine learning at scale
27mA Look Inside Applied Machine Learning
46mDev Ops meets Data Science Taking models from prototype to production with Docker
14mGetting Machine Learning applications into production using MLOps and CD4ML
35mGetting Machine Learning Models Ready For Production Using Python