Skip to content

Latest commit

 

History

History
81 lines (61 loc) · 2.62 KB

get_started.md

File metadata and controls

81 lines (61 loc) · 2.62 KB
Raw
jupyter
jupytext kernelspec
formats text_representation
ipynb,md
extension format_name format_version jupytext_version
.md
markdown
1.1
1.1.6
display_name language name
Python 3
python
python3

Getting started with Jupytext

This small notebook shows you how to activate Jupytext in the JupyterLab environment. We'll show you a few things that you can do with Jupytext and a bit of what happens under the hood.

Note: to run this notebook locally, you need to first follow the Jupytext installation instructions and activate the JupyterLab plugin. If you're on Binder, it should already work.

Enabling Jupytext in a new notebook

This notebook is brand new - it hasn't had any special extra metadata added to it.

If we want Jupytext to save files in multiple formats automatically, we can use the JupyterLab command palette to do so.

  • In the View menu, click on Activate Command Palette
  • Then type Jupytext. You should see a number of commands come up. Each one tells Jupytext to save the notebook in a different file format automatically.
  • Select Pair notebook with Markdown

That's it! If you have Jupytext installed, it will now save your notebook in markdown format automatically when you save this .ipynb file in addition to saving the .ipynb file itself.

After you've done this, save the notebook. You should now see a new file called get_started.md in the same directory as this notebook.

How does Jupytext know to do this?

Jupytext uses notebook-level metadata to keep track of what formats are paired with a notebook. Below we'll print the metadata of this notebook so you can see what the Jupytext metadata looks like.

import nbformat as nbf
from IPython.display import JSON
notebook = nbf.read('./get_started.ipynb', nbf.NO_CONVERT)
JSON(notebook['metadata'])

As you select different formats from the command palette (following the instructions above) and save the notebook, you'll see this metadata change.

That's it!

Play around with different kinds of code and outputs to see how each is converted into its corresponding text format. Here's a little Python code to get you started:

import numpy as np
import matplotlib.pyplot as plt

plt.scatter(*np.random.randn(2, 100), c=np.random.randn(100), s=np.random.rand(100)*100)

Experiment with the demo notebook!

In the "demo" folder for jupytext there is a notebook called World population.ipynb. By default, saving the demo notebook will also create many possible Jupytext outputs so you can see what each looks like and which you prefer.