JupyterLab Extension on Cloud Services

This page gives an overall account of installing NB Defense JupyterLab Extension (JLE) on different cloud platforms. Overall, the following four steps outline the installation of the JLE:

1. Choose the conda environment in which JLE is to be installed.

   conda activate ENVIRONMENT_NAME
   

2. Install the JLE using pip install command:

   pip install nbdefense_jupyter
   

3. Enable the JLE using:

   jupyter server extension enable nbdefense_jupyter
   

4. Restart the JupyterLab server with the platform specific restart command/sequence.

Please note that step 2 and step 3 will be the same on all cloud platforms, whereas step 1 and step 4 will change depending on the cloud platform.

Please follow the links for the following cloud platforms to learn more about their specific JLE installation steps:

SageMaker Notebooks

SageMaker Studio Labs

Azure Machine Learning Studio

Unsupported Cloud Platforms:

The NB Defense JLE will not work on the following cloud platforms:

1. Oracle Cloud Infrastructure (OCI): The Jupyter version running on OCI is 2.2.10 whereas the NB Defense JLE works on Jupyter version 3.

2. Google Cloud Platform (GCP): The GCP does not support third party JupyterLab extensions for both managed notebooks, and user-managed notebooks.

Other Cloud Platforms?

If you are trying to install NB Defense JLE on a cloud platform for which the installation instructions are not listed, you are welcome to open a ticket, or a PR here.