This repositry contains steps and guides on how to deploy MLFlow on OpenShift.
MLflow is an open source platform to streamline machine learning development, including tracking experiments, packaging code into reproducible runs, and sharing and deploying models. MLflow supports both traditional ML And Gen-AI applications:
- MLflow for ML: Focuses on the full lifecycle for machine learning projects, ensuring that each phase is manageable, traceable, and reproducible.
- MLflow for GenAI: Helps enhance your Agent & GenAI applications with end-to-end observability, evaluations, AI gateway, prompt management & optimization and tracking.
You can find more documentation on their official website at https://mlflow.org/docs/latest/.
We have 2 different type of MLFlow deployments:
- MLFlow with Oauth - This is the MLFlow deployment for traditional ML applications. We will be deploying MLFlow on an OpenShift cluster with an OAuth proxy i.e a reverse proxy that provides authentication with OpenShift via OAuth and Kubernetes service accounts.
- MLFlow for GenAI - This is the MLFlow deployment for GenAI applications. We will be dpeloying it on an OpenShift cluster.
In order to setup MLFlow on OpenShift you will need the following:
- OpenShift cluster with namespace available and sufficient privileges/memory to deploy
- PostgreSQL database - for storing MLFlow metadata
- S3 bucket - for storing the MLFLow artifacts such as model training files (CSVs)