An example Jupyter Notebook#

This notebook is a demonstration of directly-parsing Jupyter Notebooks into Sphinx using the MyST parser.[^download]



To build documentation from this notebook, the following options are set:

myst_enable_extensions = [
myst_url_schemes = ("http", "https", "mailto")


As you can see, markdown is parsed as expected. Embedding images should work as expected. For example, here’s the MyST-NB logo:

![myst-nb logo](../img/unitary_fund_logo.png)

myst-nb logo

By adding "html_image" to the myst_enable_extensions list in the sphinx configuration (see here), you can even add HTML img tags with attributes:

<img src="../img/unitary_fund_logo.png" alt="logo" width="200px" class="shadow mb-2">

Because MyST-NB is using the MyST-markdown parser, you can include rich markdown with Sphinx in your notebook. For example, here’s a note admonition block:


Wow, a note! It was generated with this code (as explained here):

**Wow**, a note!

If you wish to use “bare” LaTeX equations, then you should add "amsmath" to the myst_enable_extensions list in the sphinx configuration. This is explained here, and works as such:

\frac {\partial u}{\partial x} + \frac{\partial v}{\partial y} = - \, \frac{\partial w}{\partial z}

2x - 5y &=  8 \\
3x + 9y &=  -12
(1)#\[\begin{equation} \frac {\partial u}{\partial x} + \frac{\partial v}{\partial y} = - \, \frac{\partial w}{\partial z} \end{equation}\]
\[\begin{align*} 2x - 5y &= 8 \\ 3x + 9y &= -12 \end{align*}\]

Also you can use features like equation numbering and referencing in the notebooks:

$$e^{i\pi} + 1 = 0$$ (euler)
(2)#\[e^{i\pi} + 1 = 0\]

Euler’s identity, equation (2), was elected one of the most beautiful mathematical formulas.

You can see the syntax used for this example here in the MyST documentation.

Code cells and outputs#

You can run cells, and the cell outputs will be captured and inserted into the resulting Sphinx site.

__repr__ and HTML outputs#

For example, here’s some simple Python:

import matplotlib.pyplot as plt
import numpy as np
data = np.random.rand(3, 100) * 100
data[:, :10]
array([[96.33715155, 36.200233  , 83.17308466, 68.26178074, 46.18793313,
        23.12945   , 24.41389046, 29.42126059,  5.49089622, 73.84335662],
       [67.66519549, 84.89097832, 39.05727936, 90.06243558, 55.89425126,
        80.8049608 ,  3.28922145, 14.78776169, 16.44944353, 10.44046724],
       [ 7.30319043, 70.90110657, 99.81211384, 90.55839172, 71.62617594,
        38.72428369, 33.8185647 , 23.82872642, 42.46759169, 50.35609837]])

This will also work with HTML outputs

import pandas as pd
df = pd.DataFrame(data.T, columns=['a', 'b', 'c'])
a b c
0 96.337152 67.665195 7.303190
1 36.200233 84.890978 70.901107
2 83.173085 39.057279 99.812114
3 68.261781 90.062436 90.558392
4 46.187933 55.894251 71.626176

as well as math outputs

from IPython.display import Math
Math(r"\sum_{i=0}^n i^2 = \frac{(n^2+n)(2n+1)}{6}")
\[\displaystyle \sum_{i=0}^n i^2 = \frac{(n^2+n)(2n+1)}{6}\]

This works for error messages as well:

print("This will be properly printed...")
This will be properly printed...
NameError                                 Traceback (most recent call last)
Cell In[4], line 2
      1 print("This will be properly printed...")
----> 2 print(thiswont)

NameError: name 'thiswont' is not defined


Images that are generated from your code (e.g., with Matplotlib) will also be embedded.

fig, ax = plt.subplots()
ax.scatter(*data, c=data[2])
<matplotlib.collections.PathCollection at 0x7f4c8047c850>

Thumbnail for the Notebook#

To add a thumbnail for an example notebook, first add the thumbnail image file to docs/source/_thumbnails. Next, modify the docs/source/ to include the example and thumbnail in the nbsphinx_thumbnails dictionary at the end of the file. The sample below contains both a generic template and an actual example.

nbsphinx_thumbnails = {
    "examples/{}": "_static/{THUMBNAIL_FILENAME_WITH_EXTENSION}",
    "examples/hamiltonians": "_static/vqe-cirq-pauli-sum-mitigation-plot.png"