Predict Server
Run
-
docker build -t test-deploy-server:1 .
-
docker run -p 80:80 test-deploy-server:1
TODO
- Add Kubernetes Helm charts to deploy.
Potential Improvements-
- Upload predict function to a pickle and upload it to repository.
- Deploy model to kubernetes using k8 wrapper/kServe