DevMLOps at Scale
summary
A concrete enterprise pipeline gets built and shown end to end: predict wine quality from 12 features using a DBC-managed CSV, a tf-x example/statistics/schema flow, and ml flow for experiments, packaging and deployment. The pipeline is repeatable scalable and trackable (CLI, UI, hashes) and pushes containers to SageMaker / Azure ml / ECR. A 20-line tf-x pipeline generates examples to training to a pusher and the deployed endpoint returns a numeric quality (example 0.6), proving it works.
AI-written summary · about
chapters (5)
related talks
53mExperience with an end-to-end MLOps pipeline: Practical insights and 3D architectural perspectives
25mBuilding Scalable MLOps Pipelines That Actually Work
24mFrom Model to Production: Automating the ML Lifecycle with CI/CD
45mMLOps and ML-Flops: going from dev to prod
13mMLOps at Scale
56mDevOps for Machine Learning
10mFrom Model to Production: Building Scalable MLOps Pipelines
34mImplement MLOps practices with Amazon SageMaker