A Beginner’s Guide to Using Jupyter Notebooks Effectively

What is a Jupyter Notebook?

It is a tool for combining code, text and visualisations into a single document, allowing the user to iterate on ideas quickly.

On the official Jupyter website, the Jupyter Notebook is described as the original web application for creating and sharing computational documents. It offers a simple, streamlined, document-centric experience. Stripping it back, as the title suggests, the Jupyter Notebook is a notebook for storing and sharing ideas. What makes it powerful is in its ability to allow for rich text elements to be written and code to be executed all in one place.

Put simply, a Jupyter Notebook allows you to try out coding ideas fast, and document your thinking in plain text right next to your code.

Why Jupyter Notebooks?

Interactive Code Execution and Debugging

The clean interface induces a methodical approach to structuring your code, making debugging more straightforward.

A Jupyter Notebook is split into cells that can either store markdown or code. Code can be written in each cell and executed in real-time, with the output of each coding cell appearing immediately below each cell. This is incredibly useful for debugging since you only have to search each cell for a bug rather than investigating the whole script.

These cells combined make up your Notebook— but like any notebook, your notebook can be as formal or informal as you would like, making Jupyter Notebooks perfect for failing fast and iteratively improving your projects.

Note: I have used Jupyter Notebooks in coding interviews. When the pressure is on, I’ve found that they help to keep my thought processes logical and methodical.

Clean and Exportable Format

Markdown has an incredibly powerful syntax that allows you to create beautifully constructed documents that can stand alone to deliver and help communicate the value of your project.

Prior to discovering Jupyter Notebooks, I would spend lots of time flitting back and fourth between my coding environment and a word document that I was using to write up my findings. With Jupyter Notebook there is no need — the code and the write up are intertwined. Not only does this help save you time, it’s much easier for someone else to review the logic of your code and support on your project.

5 Steps to Mastery

Step 1: Install

To install Jupyter, you must have Python installed. To check whether Python is installed, you can type python --version into a terminal window. If installed, you will see the python version installed on your machine ("Python 3.6.7 for example").

If you don't see a version, you will need to install Python by running the following command:

brew install python

This will install the latest version of Python using the Homebrew package manager. If you don't have Homebrew installed on your machine, you can find further instruction here.

Once Python has successfully installed, you can install Jupyter Notebooks but running a pip command (see more about pip here):

pip install jupyter

Once the install is complete, run Jupyter by typing

jupyter notebook

into the terminal window. From here, a new web browser window will pop up with the Jupyter Notebook interface!

Step 2: Virtual Environments

Virtual environments allow a user to isolate specific Python environments, allowing for separate dependencies and packages to be stored for different projects. This helps solve issues that arise when you require a specific version of a package for one project but a different version of the same package for a different project.

Using a virtual environment with Jupyter requires an additional step to using one with pure Python. Use the process below to make your virtual environment compatible with your Jupyter Notebook setup.

Create a virtual environment using the venv module:

python -m venv myJupyterEnv

Activate your virtual environment:

source myJupyterEnv/bin/activate

Once activated, install the ipykernel package by running:

pip install ipykernel

Create a kernel for the virtual environment:

python -m ipykernel install --user --name=myJupyterEnv

Navigate to Kernel on the navigation bar and hover over Change kernel on the dropdown. Your virtual environment should now be visible for you to select.
Once the virtual environment is synced to the kernel, packages can be installed to the virtual environment with conventional pip install commands in the terminal.

Step 3: Keyboard Shortcuts

You can see the available keyboard shortcuts for Jupyter Notebooks by highlighting Help on the navigation bar and selecting Keyboard Shortcuts.

When working in Jupyter Notebooks, you will quickly realise that there are two modes of operation, either you are in Command Mode (entered by pressing ESC key) or Edit Mode (entered by pressing ENTER key).

Edit Mode holds the more conventional editing shortcuts that you use day to day in word documents. Command Mode is the real time saver and is used to help you navigate between cells and execute code blocks.

My most used shortcuts are as followed:

↩: enter edit mode

⇧↩: run cell, select below

⌃↩: run selected cells

Y: change cell to code

M: change cell to markdown

↑: select cell above

↓: select cell below

A: insert cell above

B: insert cell below

X: cut selected cells

C: copy selected cells

V: paste cells below

D,D: delete selected cells

H: show keyboard shortcuts

0,0: restart the kernel (with dialog)

Step 4: Markdown Syntax

The ability to use Markdown syntax to quickly beautify your documents in a consistent way is one of the best things about Jupyter in my opinion.

From Step 3, we already have seen how easy it is to switch a cell from being a coding block to a markdown cell that you can populate with free text and images. If you are familiar with the shortcuts that are available in Notion to change the format of a page, then you will have no time getting up to speed with the syntax required to quickly edit markdown cells robustly.

The syntax that I've found most useful up to this point is highlighted below (you can find a more complete list here):

# Heading

## Sub-heading

### Sub-sub-heading

- Unordered list

1. Ordered list first item

1. Ordered list second item

_Italics_

__Bold text__

> Quoting text

`Nested code`

[Displayed text for link](url)

Step 5: Download As…

With Jupyter Notebooks, there’s no need for file conversion, you can save as PDF, Latex or whatever other format you think is appropriate to distribute your work. Just navigate to Download as dropdown from the File button.

You can also convert the .ipynb file to a .py file, effectively converting your document into a useable script. From here you can do many things, create an executable so at the click of a button, someone can run your code? No problem.

Conclusion

The above steps are the building blocks that are required to master Jupyter Notebooks. Of course, to deliver real value to your stakeholders you will need to master more than just how to drive your IDE effectively however if used correctly, Jupyter Notebooks offer a solid platform for which users can iterate quickly, document clearly and progress projects that are free of bugs and easy to collaborate on.

How to Use Jupyter Notebooks: The 5 Steps to Mastery

A Beginner’s Guide to Using Jupyter Notebooks EffectivelyWhat is a Jupyter Notebook? It is

The all-in-one writing platform.

Write, publish everywhere, see what works, and become a better writer - all in one place.