Container workflows for data science and ML
summary
Claims OpenShift is a powerhouse for ML and shows concrete tooling and demos: uses OpenShift and radanalytics.io to spin up an app-scoped Spark cluster, uses source-to-image (s2i) and mybinder.org patterns to build a notebook into a container, runs a k‑means training notebook, packages the model behind a REST endpoint with a validator, and returns correct integer labels and a descriptive error for a 3‑D input. Also covers monitoring model drift via platform metrics and versioned data with Pachyderm.
AI-written summary · about
chapters (7)
- 0:00Start
- 10:26Immutable images
- 11:03Stateless microservices
- 11:43Declarative app configuration
- 12:37Integration and deployment
- 21:23radanalytics.io
- 22:03Open Data Hub
related talks
01h 23mIntelligent applications in containers
28mDevOps for Data Science With Kubernetes
36mData Science Meets DevOps: Gitops with OpenShift
33mCompliant Data Management and Machine Learning on Kubernetes
17mProductionizing Data Science Workloads
35mCase Study: AI-as-a-Service on Kubernetes at Scale and In Production
36mMLOps from the edge to on demand visualization
18mBeyond Inference: Bringing ML into Production