Jupyter Notebook

Jupyter Notebook#

Why use Jupyter?#

Jupyter is an open-source, web-based interface for interactive computing. Jupyter Notebook is Jupyter’s document format that can store live code, equations, visualizations, and text. These documents can be shared for reproducible research and published online. This book you are reading is written in Jupyter Notebook.

Jupyter supports many programming languages, including Python, Julia and R. Since the primary language of this training is Python, we will use Jupyter for it.

Install Jupyter#

Jupyter Notebook#

If you have followed through the previous section, you should have already installed Jupyter Notebook. If not, you can install it by

# activate your conda environment

# install jupyter notebook
mamba install jupyter notebook

You can launch Jupyter Notebook by

jupyter notebook

The command line will print a URL that you can open in your web browser to access. If the web browser does not open automatically, you can hold down Ctrl and click on the URL, or copy and paste the URL into your web browser.

This is the classical Jupyter Notebook interface.

JupyterLab#

Jupyterlab is the modern web interface for Jupyter. It is more powerful by allowing multiple notebooks to be edited in the same browser window. You will find it much more convenient to use.

You can install JupyterLab by

mamba install jupyterlab

You can launch JupyterLab by

jupyter lab

Tour of Jupyter Notebook and JupyterLab#

The official Jupyter website contains two brief tours:

Please visit each of them. After the page loads, you will see a message at the lower right corner, saying “Try the Notebook Tour” or “Try the Welcome Tour.”. Click “Start Now” to begin the tour.

Shortcuts#

Learning to use shortcuts can significantly increase your productivity! Here are some shortcuts to start with:

  1. When you are editing a cell, press esc to exit edit mode.

  2. When you are NOT editing a cell, you can use arrow keys to move around. Alternatively, you can use j and k to move between cells. j and k reduces the travel distance of the hand compared to arrow keys.

  3. To edit a cell, move the focus there and press Enter.

  4. You can insert a cell using b (below) or a (above).

  5. You can delete a cell using dd.

  6. You can run a cell using Shift+Enter. This will run the cell and move the focus to the next cell. If this is the last cell, a new one will be inserted.

  7. Merge a cell with the cell below it using Shift+M.

  8. To split a cell into two, move the focus to the line where you want to split, then press Ctrl+Shift+Minus(-). Think of that - as a separator.

  9. To change a cell from code to markdown, press m. To change it back to code, press y.

Advanced Features#

Magic Commands#

Magic commands are a special commands to make life easier.

Magic commands in Jupyter are special commands that start with % (line magics) or %% (cell magics). They provide quick access to various features and can significantly enhance your workflow. Here are some commonly used magic commands:

Line Magics (%)#

  1. %timeit: Time the execution of a single line of code

    %timeit [1, 2, 3, 4, 5].sort()

  2. %matplotlib inline: Display matplotlib plots inline in the notebook

    %matplotlib inline
import matplotlib.pyplot as plt
plt.plot([1, 2, 3, 4])
plt.show()

  3. %who: List all variables in the current namespace

    %who

  4. %load: Load code from an external file

    %load my_script.py

Cell Magics (%%)#

  1. %%time: Time the execution of an entire cell. This is useful when you want to understand the performance of a code block

    %%time
import time
for i in range(1000000):
    _ = i**2

  2. %%writefile: Write the contents of a cell to a file

    %%writefile example.py
def hello():
    print("Hello, World!")

  3. %%html: Render the cell as HTML

    %%html
<h1>This is a header</h1>
<p>This is a paragraph.</p>

You can get a list of all available magic commands by running %lsmagic. To get help on a specific magic command, you can use %magic_command? (e.g., %timeit?).

Magic commands are a powerful feature of Jupyter that can help you write more efficient and interactive notebooks. Experiment with different magic commands to find the ones that best suit your workflow!

Markdown and LaTeX#

Jupyter supports Markdown for rich text formatting and LaTeX for mathematical equations:

# Header 1
## Header 2
- List item 1
- List item 2

Inline equation: $E = mc^2$
Block equation:
$$
\frac{\partial f}{\partial x} = 2x
$$

Exporting Notebooks#

Jupyter notebooks can be exported to various formats. One can use the menu “File” - “Save and Export Notebook As” to export the notebook from the .ipynb format to others.

Notebooks can also be converted with commands:

jupyter nbconvert --to pdf your_notebook.ipynb
jupyter nbconvert --to html your_notebook.ipynb

Jupyterlab Extensions#

In addition to the standard features, Jupyterlab can be extended by using third-party extensions. These third-party extensions provide features such as code formatting, integration with Google Drive, or integration with GitHub.

Extensions can be installed by clicking on the “puzzle piece” icon. Upon the first use, one will be warned about the risks of using third-party extensions. After acknowledging the risks, one can use the search feature to find extensions, or browse all of them.

There are some good extensions to begin with, but one may install them as needed, becasue having to load many extension may slow the startup of Jupyterlab and increase memory consumption on the client browser.

Try searching for “formatter” and installing jupyterlab-code-formatter. When done, in a new notebook window, a “format notebook” icon will appear. To use it, highlight the code to reformat, and click on the formatter.

Other useful extensions to consider:

  • jupyterlabcodetoc: for adding a table of contents, which helps navigate in a large but organized notebook

  • jupyterlab-git: for git integration within Jupyterlab

Themes#

JupyterLab supports both light and dark themes just like your newer phone and computer operating system. You can change the theme by going to Settings > Theme in the JupyterLab menu.

Using a dark theme can be easier on the eyes, especially when working in low-light environments.

Setting Password#

You may have noticed that Jupyter generates a random session key each time you start it. That looks like

http://localhost:8888/lab?token=85f1ba706b2151a5d8a9c0b6be01b86db62854935798ab27

If you are using Jupyter on a local WSL, it may be helpful to use your own password, so that you can head directly to the browser if the automatic pop-up doesn’t work. Also, setting a password for your Jupyter server is an important security measure, especially if you’re running Jupyter on a network where others might access it.

Here’s how to set up a password with a recent version (>v5.0) of Jupyter Notebook:

  1. First, generate a config file if you don’t already have one:

    jupyter notebook --generate-config

  2. Then, set up a password:

    jupyter notebook password

    You’ll be prompted to enter and verify your password.

  3. This will create a hashed password in your Jupyter config file. You can now start Jupyter normally, and you’ll be required to enter this password to access your notebooks.

Remember, it’s crucial to choose a strong, unique password to protect your Jupyter environment and any sensitive data in your notebooks.

When running Jupyter on a shared or public network, consider additional security measures like using SSL encryption or running Jupyter behind a reverse proxy.

For more advanced topics, refer to the official Jupyter documentation and community resources.

Contents