GitOps for Machine Learning Pipelines
summary
Shows that GitOps applies beyond CI/CD by wiring GitHub, Kubernetes and Kubeflow to run, version and promote ML work: experiments are declared as PRs, 100 training runs on Tiny ImageNet are orchestrated and compared via TensorBoard, data and model snapshots use DVC for data/model versioning, and customized Kubeflow components are deployed through Argo CD deployment to a cluster (full Kubeflow deploy time ~30–45 minutes).
AI-written summary · about
chapters (4)
related talks
57mDevOps for Machine Learning
05mCI/CD Pipelines and Machine Learning
56mDevOps for Machine Learning
26mGitOps Lifecycle Management
32mBridging DevOps and MLOps: Unifying Pipelines with KitOps and GitOps
24mFrom Model to Production: Automating the ML Lifecycle with CI/CD
24mEmpowering the GitOps Future with AI
36mData Science Meets DevOps: Gitops with OpenShift