Using a custom notebook image

As you have seen in Chapter 5, Data Engineering, JupyterHub allows you to spin up Jupyter Notebook-based development environments in a self-service manner. You have launched the Base Elyra Notebook Image container image and used it to write the data processing code using Apache Spark. This approach enables your team to use a consistent or standardized development environment (for example, same Python versions and same libraries for building code) and apply security policies to the known set of software being used by your team. However, you may also want to create your own custom images with a different set of libraries or a different ML framework. The platform allows you to do that.

In the following subsection, you will build and deploy a custom container image to be used within your team.

Building a custom notebook container image

Let's assume that your team wants to use a specific version of the Scikit library along with some other supporting...