JupyterHub | Kale | Kubeflow Pipelines | on local Kubernetes cluster - konsloiz/masters-thesis GitHub Wiki
Installation Steps
1. JupyterHub
Follow this guide
Recommended Helm JupyterHub chart: helm install jhub-datascience jupyterhub/jupyterhub --version=v0.11.0 -f config.yaml -n jupyter --timeout 180s
To allow git-cloning, run in a terminal from a notebook server:
git config --global http.sslverify false
Important: Due to Kale compatibility reasons (can run on "jupyterlab>=2.0.0,<3.0.0"), install any Helm JupyterHub chart >=0.10.0,<=0.11.0.
2. Kale
For a server-notebook standalone installation, open a terminal inside JupyterLab and enter the following commands:
npm config set strict-ssl false
npm install --global yarn
yarn config set "strict-ssl" false
pip install kubeflow-kale
jupyter labextension install kubeflow-kale-labextension
jupyter lab build and refresh
3 Kubeflow Pipelines
Follow this guide
export PIPELINE_VERSION=1.7.0
kubectl apply -k "github.com/kubeflow/pipelines/manifests/kustomize/cluster-scoped-resources?ref=$PIPELINE_VERSION"
kubectl wait --for condition=established --timeout=60s crd/applications.app.k8s.io
kubectl apply -k "github.com/kubeflow/pipelines/manifests/kustomize/env/dev?ref=$PIPELINE_VERSION"
kubectl port-forward -n kubeflow svc/ml-pipeline-ui 8080:80
Open the Kubeflow Pipelines UI at http://localhost:8080/