Jupyter Notebooks: Interactive Computing in Your Browser

Before You Start

You should know: - How to create a virtual environment, activate it, and install libraries using pip or conda.

You will learn: - The concept of β€œliterate programming” and why notebooks revolutionized spatial data science. - How to install, launch, and navigate the Jupyter interactively. - The fundamental difference between a Code cell and a Markdown cell.

Introduction

So far, you’ve learned how to establish a virtual environment sandbox and populate it with geographic and mathematical libraries. The question remains: where do you actually write the Python code?

The traditional way is to open a plain text file, write hundreds of lines of code, save it as script.py, and run the entire file from the command line. If an error occurs on line 499, the program crashes, and you must run the entire file again from the top.

If you are just exploring data or making a quick map, rerunning the entire file is agonizing. This is where Jupyter Notebooks enter the picture.

Jupyter allows you to chop your Python code into tiny blocks called β€œcells,” and execute them one by one. The output (a map, a graph, or an error) appears immediately beneath the cell. Because it allows you to dynamically write code, see the result, and iterate instantly, Jupyter has become the undisputed standard environment for scientific computing.

Literate Programming

A Jupyter notebook is not just a code environment; it is a document.

Jupyter is built on the philosophy of literate programming. In a notebook, you can create a block of code, followed by a block of beautifully formatted text, followed by an interactive chart. You are writing an executable narrative.

In fact, almost the entirety of this Computational Geography platformβ€”including pathways, assays, and problem setsβ€”was written, executed, and rendered out of Jupyter Notebooks!

Setting Up Jupyter

Jupyter isn’t a program you download from a website; it is actually a Python package you install into your virtual environment!

Open your terminal, ensure your virtual environment is active, and install it:

Using pip:

pip install jupyter

Using conda:

conda install -c conda-forge jupyterlab

Launching Your First Notebook

To start writing, you must tell your computer to spin up the Jupyter server.

In your terminal, navigate (cd) to your project directory. Then, type:

jupyter notebook

(If you installed JupyterLab, type jupyter lab instead).

Your terminal will lock up and start printing server logs. Then, magically, your default web browser will open a new tab containing the Jupyter interface.

[!NOTE] Even though you are in a web browser, Jupyter is not connecting to the internet. It is rendering a local interface that talks directly to your computer’s hardware. If you disconnect from Wi-Fi, Jupyter works perfectly!

Working With Cells

Once inside Jupyter, click New > Notebook. You are now looking at an empty document with a single gray box. That box is a Cell.

Cells come in two flavors:

1. Code Cells

By default, the cell expects Python. Type a basic math equation into the cell, like 2 + 2. To execute the cell, press Shift + Enter. It will compute the output, print 4 beneath it, and spawn a new empty cell.

2. Markdown Cells

You can change a cell from β€œCode” to β€œMarkdown” using the dropdown menu at the top of the screen. Markdown cells are designed for text. You can use hashtags (#) for headers, asterisks (*) for bullet points, and dollar signs ($) to write LaTeX math equations. Change a cell to Markdown, type # My First Notebook, and hit Shift + Enter. It will instantly render into a beautiful header.

Stopping Jupyter

Jupyter is a running background server. If you simply close your web browser tab, the server continues to run invisibly on your computer, consuming memory.

To properly shut down Jupyter: 1. Save your notebook in the browser. 2. Go back to your terminal window (which should still be printing out server connection logs). 3. Press Ctrl + C twice in rapid succession.

This sends a β€œkill” signal to the server. Your terminal will return to a blinking cursor.

Verify Your Work

  1. Activate your virtual environment and install Jupyter.
  2. Launch jupyter notebook.
  3. Create a new notebook document.
  4. Execute print("Hello Geography!") in a Code cell.
  5. Create a Markdown cell below it and write a brief sentence.
  6. Safely shut down the Jupyter server using Ctrl + C in the terminal.

You now possess a fully functional, interactive computing landscape.