Create notebook

After the notebook server has started, we can create our first notebook.

Create a notebook

In your standard browser you should see the notebook dashboard with the New menu on the right. All notebook kernels are listed in this menu, but initially probably only Python 3.

After you have selected New ‣ Python 3, a new notebook Untitled.ipynb will be created and displayed in a new tab:

Renaming the notebook

Next you should rename this notebook by clicking on the title Untitled:

The notebook user interface

There are two important terms used to describe Jupyter Notebooks: cell and kernel:

Notebook kernel

Computational engine that executes the code contained in a notebook.

Notebook cell

Container for text to be displayed in a notebook or for code to be executed by the notebook’s kernel.

Code

contains code to be executed in the kernel, and the output which is shown below.

In front of the code cells are brackets that indicate the order in which the code was executed.

In [ ]:

indicates that the code has not yet been executed.

In [*]:

indicates that the execution has not yet been completed.

Warning

The output of cells can be used in other cells later. Therefore, the result depends on the order. If you choose a different order than the one from top to bottom, you may get different results later when you e.g. select Cell ‣ Run All.

Markdown

contains text formatted with Markdown, which is interpreted as soon as Run is pressed.

What’s an ipynb file?

This file describes a notebook in JSON format. Each cell and its contents including pictures are listed there along with some metadata. You can have a look at them if you select the notebook in the dashboard and then click on edit. For example the JSON file for my-first-notebook.ipynb looks like this:

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# My first notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello World!\n"
     ]
    }
   ],
   "source": [
    "print('Hello World!')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

Save and checkpoints

When you click on Save and Checkpoint, your *.ipynb file will be saved. But what is the checkpoint all about?

Every time you create a new notebook, a file is also created, which usually automatically saves your changes every 120 seconds. This checkpoint is usually located in a hidden directory called .ipynb_checkpoints/. This checkpoint file therefore enables you to restore your unsaved data in the event of an unexpected problem. You can go back to one of the last checkpoints in File ‣ Revert to Checkpoint.

Tips and tricks

1. Give the notebook a title (# MY TITLE) and a meaningful foreword to describe the content and purpose of the notebook.

2. Create headings and documentation in Markdown cells to structure your notebook and explain your workflow steps. It doesn’t matter whether you do this for your colleagues or for yourself in the future.

3. Use Table of Contents (2) from the List of extensions to create a table of contents.

4. Use the notebook extension setup.

5. Use snippets from the list of extensions to add more frequently used code blocks, for example typical import instructions, easy to insert.