leiloong / jupytemplate

Templates for jupyter notebooks

Jupyter template

A simple template for jupyter notebooks.

The extension sets up any new Jupyter Notebook with a conventional and general-purpose template to shape Data Science analysis.

The template includes conventional sections, like Data Import, Processing and References, as well as code to perform common operations, like importing and configuring charting libraries.

Moreover, it prompts for a meaningful name whenever you try and save a notebook called 'Untitled'.

You find this annoying? Don't worry, you can disable this one.

Motivation

Jupyter notebooks are awesome tools: they enable fast prototyping and ease result sharing. However, due to their flexibility, they are prone to be abused.

In order to help Data Scientists keep their notebooks clean, a reasonably flexible yet conventional template may help. Moreover, the template is also a productivity tool, speeding up common setup, such as library import and configuration.

Quick start

We assume Jupyter notebook is already installed in your environment. However, even if this is not tha case, don't worry: jupytemplate declares Jupyter notebook as a dependency, thus any package manager, like pip, will install it for you.

It is not mandatory, but you can install the full set of Jupyter extensions.

pip install jupyter_contrib_nbextensions
jupyter contrib nbextension install --user

Feel free to visit their repository for more information.

Now you can install the package:

pip install jupytemplate

Then, you have to install the javascript files from the Python package in a conventional jupyter directory:

jupyter nbextension install --py jupytemplate --sys-prefix

Finally, you may want to enable the extension:

jupyter nbextension enable jupytemplate/main --sys-prefix

You can easily enable, disable or configure the extension by using the nbextension_configurator server extension, as shown below.

Editing the template

Template location can be found by running:

import jupytemplate
print(jupytemplate.get_template_path())

Of course, you can edit the template as you like, in order to adapt it to your own needs, but keep the file name template.ipynb.
After editing the template, run:

jupyter nbextension install --py jupytemplate --sys-prefix
jupyter nbextension enable jupytemplate/main --sys-prefix

to make changes effective.

References

Please consider reading the following resources for a more comprehensive understanding:

• Emanuele Fabbiani, Stop copy-pasting notebooks, embrace Jupyter templates
• Will Kohersen, Set Your Jupyter Notebook up Right with this Extension
• Will Kohersen, How to Write a Jupyter Notebook Extension
• Will Kohersen, Setup extension