Extensions

You can extend the use of Jupyter notebooks by importing external extensions, building your own magics, or even creating your own kernels.

External extensions and libraries

Callysto has our own collection of importable extensions, nbplus.

You can also use pip install !pip3 install foo --user where foo is the library you want to install. Generally this is discouraged, we want to have programs installed on the docker image of the server rather than in individual notebooks. This avoids having messy user facing installs and helps maintain consistency across the notebooks. However there are exceptions, if you are unsure ask the developer channel in slack.

We discourage the use of the extensions in the nbextensions package as many of them are unstable and may cause issues with the server. We also discourage use of plotting libraries outside of matplotlib/plot.ly/D3/mathbox, more about this can be found in the plotting section of the manual. For tutorials and example implementations of standardized curriculum notebook tools, see https://github.com/callysto/shorts/.

Requesting an extension added to the hub

If there is an extension you would like to add to the server please create a notebook demonstrating the basic use cases for the extension and then post the request on the developers slack channel.

Magics

Magics allow you to use many different programming languages in Jupyter cells as well as extend the functionality of Jupyter by bringing useful functionality into the cell. Clone the Magics Guide notebook or view it with nbviewer for more information on using and creating magics.

Kernels

The primary languages used by Callysto are Python 3, Javascript and HTML/CSS. All notebooks you create should be based on a Python 3 kernel (there may be exceptions). You can bring in additional kernels using magics (see subsection above).

Last updated