Data Science from Research to Production with Jupyter, Kubeflow & Nuclio
Ai-Week (17.11.2019)
3 hour workshop, showing how to use Nuclio & MLRun with Kubeflow to deploy jupyter notebooks as serverless functions or pipelines without leaving the jupyter environment.
Preperations
Install Nuclio [website|github]
- Locally on docker:
docker run -p 8070:8070 -v /var/run/docker.sock:/var/run/docker.sock -v /tmp:/tmp nuclio/dashboard:stable-amd64 - On Kubernetes (Or local minikube) / AKS / GCP: Nuclio installation notes
Jupyter integration
pip install nuclio-jupyter
Install MLRun [github]
pip install mlrun