mirror of
https://github.com/Brandon-Rozek/website.git
synced 2024-11-26 18:13:59 -05:00
21 lines
1.2 KiB
Markdown
21 lines
1.2 KiB
Markdown
|
---
|
||
|
title: "Live Documentation"
|
||
|
date: 2019-09-27T23:07:19-04:00
|
||
|
draft: false
|
||
|
---
|
||
|
|
||
|
This blog post is mostly for one of my teams in which I use Jupyter Notebooks for documentation. Perhaps after reading this post, you the reader can understand why it might be beneficial to use Jupyter Notebooks as a form of documentation.
|
||
|
|
||
|
## Why?
|
||
|
|
||
|
So why Jupyter Notebooks?
|
||
|
|
||
|
- Follows the literate programming approach. You can write text explaining a feature and then immediately show code and it's result.
|
||
|
- It's modifiable. If your user wants to play around with the documentation, the environment is set up for them to do so.
|
||
|
- It's exportable. Let's say another user doesn't want to bother setting it up. Well it's super simple to just export the notebook as a PDF and send that to them instead.
|
||
|
|
||
|
## Setting up
|
||
|
|
||
|
Jupyter Notebooks are part of the [Project Jupyter](https://jupyter.org/) suite of products. You can install it via a `pip` package, but it is more commonly installed via the [Anaconda Distribution](https://www.anaconda.com/)
|
||
|
|
||
|
Once you have that installed, run `jupyter lab` in the directory that you wish to execute code from. You might need to be in the `bash` shell for this to work since the installer modifies those environmental variables.
|