An MLOps pipeline leveraging the cloud-native Kubeflow project for model training and registration with MLflow. GitOps practices are implemented using ArgoCD to deploy a custom Kubeflow Helm chart to a Kubernetes cluster on DigitalOcean.
Application | Git Repository | Version |
---|---|---|
Kubeflow | https://github.com/mahdikhashan/kubeflow-helm-chart | 0.0.1 |
MLFlow | https://github.com/mahdikhashan/mlflow-helm-chart | 0.0.1 |
Grafana | https://grafana.github.io/helm-charts | 8.6.4 |
Prometheus | https://prometheus-community.github.io/helm-charts | 26.0.0 |
- Setup
- Research Summary
- Learning
- Architecture
- Challenges
- Components