Overview of Jupyter Notebook

IDEs are software applications that provide software programmers with a suite of services such as coding interfaces, debugging, and compiling/interpreting. Python programmers are spoilt for choice as there are many open source IDEs for Python, including Jupyter Notebook, spyder, atom, and pycharm, and each IDE comes with its own set of features. We have used Jupyter Notebook for this book. and all the code and exercises discussed in this book can be accessed at https://github.com/PacktPublishing/Hands-On-Python-Natural-Language-Processing.

Jupyter Notebook is the IDE of choice for pedagogical purposes as it allows us to weave together code blocks, narrative, multimedia, and graphs in one flowing notebook format. It comes pre-packaged with the Anaconda Python distribution and installing it is quite simple. Please refer to the very nicely written beginner's guide, which should help you gain a basic understanding of Jupyter Notebook: https://jupyter-notebook-beginner-guide.readthedocs.io/en/latest/execute.html.

Jupyter Notebook has an .ipynb extension. In order to launch a notebook, open the terminal (if you have installed Anaconda, please use the Anaconda Prompt) and cd to the directory where the notebook is located. Run the jupyter notebook command, which will launch a starter page that lists the files stored in that location. You can either select an existing notebook to open or create a new notebook by clicking on the New button and selecting Python3, as shown in the following screenshot:

This will create a new notebook with a single cell. By default, this cell is meant for you to type your code into it. However, using the drop-down menu shown in the following screenshot, you can toggle between Code and Markdown (text):

You can either use the icons in the notebook to insert/run cells or use hot keys such as Shift + Enter to run the current cell, Ctrl + Enter to run multiple selected cells, A to insert a cell above, B to insert a cell below, and so on. Once you have completed working on the notebook, you can save it in the format of your choice by navigating to File | Download as, as shown in the following screenshot. Jupyter provides various options for you to save the file based on the requirement (although you would typically want to save it in Jupyter Notebook (.ipynb) format):

You can also access a finished notebook (a pre-existing .ipynb file) by running the jupyter notebook <filename> command. This will open the required notebook in a browser. The following are some screenshots of launching and working on a completed notebook.

The following is a screenshot of running a Jupyter Notebook cell with code:

The following screenshot shows how variables can be visualized in Jupyter Notebook inline by running the variable name. You can also see how bar plots can be rendered inline by running the barplot command:

The following screenshot shows how easily you can render a histogram or distribution plot in the Jupyter notebook and how you can add text just below the plot to explain the main points to potential readers:

The following screenshot shows how a count plot can be rendered inline and explained using rich text:

Given its powerful features and ease of use, Jupyter Notebook has become one of the most popular Python IDEs in both academia and industries. Please note that the authors have no intention of persuading you to switch to Jupyter Notebook if you are already comfortable with another IDE. However, we would very much appreciate it if you attain basic familiarity with Jupyter Notebook as the supporting code and exercises in this book have been composed as Notebook.