Once you have ML Studio installed, find out its the IP address to access it, login and start building.
kubectl get svc istio-ingressgateway -n istio-system
USER: adminPASSWORD: password
ML Studio root work directory can be access at /home/jovyan/work
inside jupyter pods and /home/work
inside other pods.
All files you would like persisted should live under this directory.
We recommend using conda
to manage all your conda environment but you are free to use others.
base
conda environment will not get persisted
Start a new terminal session under Notebooks
conda create --name py3 python=3.7 -y
Before you can use this new python enviorment in your notebooks, you need to make Jupyter aware of it first.
conda activate testipython kernel install --user --name=py3
Finally, refresh ML Studio web app for your changes to take effect.