The Teradata JupyterLab Docker image is built from the "JupyterLab Data Science Notebook" Docker image, which includes many Python, R, and Julia libraries. But for those users who wish to extend or add additional libraries to the Teradata JupyterLab Docker image, it is a fairly simple procedure. The JupyterLab Data Science Docker image uses the Conda packaging environment, so you will use the "conda install" command to install your library from conda-forge into your new Docker image.
Load the docker image into your docker repository: docker load -i teradatajupyterlabext.tar.gz
Create a simple Docker file like:
RUN conda install --quiet --yes \ 'NameOfLibFromCondaForge'
4. Then, from the directory containing your new Dockerfile, run: docker build -t mynewdockername .
This will generate a new Docker image that contains everything from the Teradata JupyterLab, teradatajupyterlabext, Docker image plus any new libraries you included. Then you simply run your new Docker image.
You can also save your new Docker image to a tar zip file to share with others by running:
docker save mynewdockername | gzip -c > mynewdockername.tar.gz
Or you can push your new Docker image to your own repository,
NOTE: Keep your Docker image current and up-to-date by periodically checking for any updates to the Teradata Jupyter Extension Docker image.