Simple repo for learning ML Ops to run a ML model as microservice using Flask

What's included:

• simple code for retraining the model
• flask API (development configuration, not production-ready yet)
• Dockerfile for building an image
• Deployment configuration for testing running the cluster in minikube (local Kubernetes)

Backlog:

• maybe move from ubuntu to lighter Docker image
• add monitoring capability (via Prometheus etc.)
• potentially add various endpoints to test serving image / audio data