Jupyter

In the last section we touched on R Markdown, which was a great introduction to interactive computing in a reproducible way. Another standard that people love to use for data analysis, are jupyter notebooks. There are a lot of different ways to interact with Jupyter Notebooks and some pros and cons to them that we'll address.

Try it in the Browser

mybinder allows you to run notebooks in the cloud and allows you to turn a git repo into a collection of interactive notebooks! Follow the link and let's test out a few different notebooks they have. Try the Classic Notebook, JupyterLab and Jupyter with R, in that order preferably.

Binder

Once you're done getting a feel for Jupyter notebooks let's add your GitHub repo to binder! Grab the binder badge to stick in your README as well. For example:

Using VS Code

1. Open up VS Code and connect to giant.
2. Make sure you have the Jupyter extension installed on the remote host.
3. Search for the command Python: Open Start Page
4. Click the link to the sample notebook and install the jupyter requirements when prompted.

When you're down playing around with the tutorial:

1. Create a new notebook in your git repo with the command Jupyter: Create New Blank Notebook.
2. Add a code block with print("Hello") and a markdown code block.
3. Commit your notebook to your git repo.

Back to Binder

First, commit your notebook. If VS Code doesn't pick up the notebook git add first-notebook.ipynb is the command. Push the changes up and let's check them out on GitHub. Note that GitHub prints it in a pretty way, but let's check out the Raw format. Note how the it's a bunch of data, and not very human readable. You'll see that this makes diffs difficult to process in the future and is a draw back of Jupyter Notebooks.

Now that you've added a Jupyter notebook to your repo try opening it in binder. This is a quick and easy way to share notebooks with collaborators.

Further Reading

• You can also use Google's flavor of Jupyter notebooks at Google Collab