iptizer/kubeflow-datascience

JupyterLab image for Kubeflow including kale.

kubeflow-datascience

This notebook is built on the jupyter/datascience-notebook and adds:

• boto3 - Access AWS through Python
• Jupyterlab Language server for R and Python

Currently NOT part of this image:

• Elyra pipeline editor
• Kale

Add to Kubeflow

Use the following command to edit the configmap:

NAMESPACE=kubeflow
CONFIGMAP=$( kubectl get cm -n $NAMESPACE -l app=jupyter-web-app -o jsonpath="{$.items[0].metadata.name}")
kubectl edit cm -n $NAMESPACE $CONFIGMAP

In data.spawner_ui_config.yamledit the list spawnerFormDefaults.image.value set the default image.

In data.spawner_ui_config.yamledit the list spawnerFormDefaults.image.options. Add/ replace the following as a list item.

docker.io/iptizer/kubeflow-datascience:latest

After that, you need to restart the rollout of the deployment:

kubectl rollout restart deploy/jupyter-web-app-deployment -n kubeflow

Image

This image is available on https://hub.docker.com/repository/docker/iptizer/kubeflow-datascience or may be pulled with the following command:

docker pull docker.io/library/iptizer/kubeflow-datascience

Or build it locally:

docker login docker.io && \
docker build . -t iptizer/kubeflow-datascience:latest && \
docker push docker.io/iptizer/kubeflow-datascience:latest