ML Ops on Google Cloud
summary
Claims a serverless, scalable ML workflow platform that logs run metadata for lineage and reproducibility and charges per executed resource use. Shows Vertex Pipelines built from reusable prebuilt components and from custom components using KFP and TFX, compiles and runs an AutoML image classification pipeline directly from a notebook and surfaces logs and artifacts in the console UI. Describes Vertex Feature Store for low latency serving and model monitoring that emits alerts and distribution stats for drift detection.
AI-written summary · about
related talks
29mML Ops on Google Cloud Platform
32mMachine Learning with Google Cloud Platform
23mBuilding Scalable, Observable MLOps Systems on Google Cloud
45mVertex AI: Pipelines for your MLOps workflows
31mProductionizing ML with ML Ops and Cloud AI
26mBuilding Portable ML Pipelines for Development and Production
31mVertex AI: Pipelines for your MLOps workflows
28mMastering MLOps: Building Future Ready ML Systems