Integrating Machine Learning - from ingestion to persistence
summary
Discusses integrating machine learning into a Big Data pipeline by using parallel ingestion and streaming tools to perform analyses earlier in the flow and reduce data movement after persistence. Describes how these techniques speed up deployment of large-scale models, drawing on practical cloud experience (AWS, Google, IBM, Oracle) with databases and cloud computing.
AI-written summary · about
chapters (11)
- 0:00ML topic intro
- 3:25ML use cases
- 10:58Data engineering challenges
- 14:10AWS data lake
- 16:43Model lifecycle
- 23:55Edge models
- 26:42Streaming and real-time ML
- 33:41Ingestion architecture
- 42:39Fraud detection
- 46:52Model deployment
- 47:15Costs and optimization
related talks
41mWant a Good Machine Learning Model? Start with Data Processing
37mBig Data Analytics, Machine Learning, Real Time Processing
01mintro video
41mInfrastructure for ML Applications
33mContinuous Delivery Principles for Machine Learning
18mContinuous deployment of Machine Learning models
27mA Look Inside Applied Machine Learning
23mAI-Driven Enterprise Intelligence