Skip to main content

Create a package from a jupyter notebook

Project description


A tiny little package to create packages from jupyter notebooks

If you find yourself copy pasting functions between jupyter notebooks, this might help.

Please note: this allows you to import functions and objects from your notebooks.

Typical Use Cases

This is very helpful when producing the same graph, running the same ETL, or executing the same ML pipeline on different datasets or different versions of the same dataset.

Make Packages from Jupyter Notebooks

In Python

You can run this in <your_notebook.ipynb> or anywhere else in the same directory


will create the needed files and directories [in the current directory] to produce a local package

Importing the Package

You can access by

import notebooks.src.<cleaned_notebook_name>.<cleaned_notebook_name>

where cleaned_notebook_name is all lower case and has spaces replaced with underscores. Please note if your notebook name has illegal characters for a python package, there will be an error.

I recommend importing your notebook as an abbreviation-- but as usual please do not import *, for the same reasons as usual.

Make sure to rerun the code if you want the package updated.

From the Command Line

You can enter

jupyckage --nb <notebook_name.ipynb>

in your terminal and the most recent saved version of your notebook will be turned into a package. You can import the package as described above in 'In Python'

Under the Hood: Package Directory Structure

With no consensus on directory structure for packages, I defaulted to a version of's suggestion.

├── src/
│   └── <notebook_name>/
│       ├──
│       └── <notebook_name>.py  # what you'll be importing
├── bin/
│   └── <notebook_name>.py #executable

I realize the import statement is hefty, but I didn't want to handle collissions in a VO.

Skip Stuff

Aything you don't want included in the package you can put below a MARKDOWN cell including the text between the quotes



Help! My code isn't updating!

Make sure that you've run your notebook and saved it. This package relies on jupyter notebook's nbconvert under the hood, which looks to the most recent checkpoint of your notebook.

Help! Why can't jupyckage do more?!?!

If you're interested added functionality, please let me know and I'll look at adding it or happily work with you so you can contribute :)

Help! Why is the import statement so verbose?!?!

I realize the import statement is hefty, but I didn't want to handle collissions in a VO. If you find this package useful and would really love an update or want to contribute, please contact me :)


Hi, I'm Kerstin Frailey. I whipped up this package quickly to make my life a bit easier. Hope it helps some other people out, too.

Here's the repo info:

Here's my contact info:

  • twitter @KEFrailey
  • blog
  • email
  • linkedin

Hope you have a nice day!

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

jupyckage-0.1.2.tar.gz (6.4 kB view hashes)

Uploaded source

Built Distribution

jupyckage-0.1.2-py3-none-any.whl (6.6 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page