Jupyter Notebook

The Jupyter Notebook is an interactive environment for writing and running code.

The notebook is capable of running code in a wide range of languages.

However, each notebook is associated with a single kernel. This notebook is associated with the IPython kernel, therefore it runs Python code.

Notebook Cells

An Notebook is a series of "cells".
The active cell has an outline around it.
Navigate to different cells with the arrow keys.
A cell can be either be code, markdown, or "raw".
- Code
  - Python
  - R
  - Julia
  - Many Others
Create cells from the Notebook menu or using keyboard shortcuts.
- h: Bring up keyboard shortcuts screen.

Modes

	Edit mode	Command mode
How to enter	`Enter`, or double-click	`Esc` or `Ctrl-m`
Cell border	Green	Blue
Action	Type code or text	Notebook-level actions

Some shortcuts

Ctrl-Enter: Run Cell
Shift-Enter: Run cell, select cell below.
Alt-Enter: Run Cell, insert below.
a : Insert cell (above)
b : Insert cell (below)
dd : Delete cell (the same as VIM)
Ctrl-s: Save Notebook
i: Interrupt kernel (press twice).
0: Restart kernel (press twice).

Features

1. Nice Errors



In [ ]:

    
#1 + foos

2. Inline Documentation



In [ ]:

    
open?

3. Tab Completion (predictive and helpful!)

Type o then tab and you should see a drop-down menu of choices.



In [ ]:

    
#o     # Always press the tab! Tab is the same as Unix. It works a little like predictive text messaging if you are unfamilar with it.

4. Run Shell Commands with "!"



In [ ]:

    
!ls  # ls is a Unix command for list. Here we are listing all files within out current Jupyternotebook directory.

5. Remote Operations

You can run code both locally and remotely.
For example, you can connect to HPC via SSH protocol and run your code there.

6. Embed Images or Videos



In [ ]:

    
from IPython.core.display import Image 
Image(url='http://python.org/images/python-logo.gif')



In [ ]:

    
Image(url='https://jupyter.org/assets/main-logo.svg',width=200)



In [ ]:

    
from IPython.lib.display import YouTubeVideo
YouTubeVideo("7qym7b-qvkE")

7. Load Remote Code - This is Very Cool and helpful!

You can pull up remote code using loadpy

loadpy http://matplotlib.org/mpl_examples/mplot3d/2dcollections3d_demo.py



In [ ]:

    
#loadpy http://matplotlib.org/mpl_examples/mplot3d/2dcollections3d_demo.py

You can also notice that probably the best feature of the notebook is inline figures.

The file itself is simply a JSON file which can be shared the same as any other file.

It is possible to save a notebook in various formats via the File->Download As menu.
- HTML
- Markdown
- Python script
- PDF
- etc.

NBviewer - view (but not execute) Jupyter Notebooks

Cloud services:
- Wakari
- Binder
- SageMath

References

This Notebook was build upon materials from the Unidata Python Workshop https://github.com/Unidata/unidata-python-workshop

Jupyter Notebook

Notebook Cells

Modes

Some shortcuts

Features

1. Nice Errors

2. Inline Documentation

3. Tab Completion (predictive and helpful!)

4. Run Shell Commands with "!"

5. Remote Operations

6. Embed Images or Videos

7. Load Remote Code - This is Very Cool and helpful!

Sharing Notebooks

References