Jupyter

This page is meant to be a startup guide for using Jupyter Notebooks with Python. It assumes you have installed Anaconda from https://www.anaconda.com/. Most of this guide was written running Python 3.9 and Jupyter Notebooks 6.4.12.

Starting Up

• To start Jupyter Notebooks with Anaconda:
  • On Windows, go to the Anaconda folder in the Start Menu or open the Anaconda Navigator and start Jupyter Notebooks from there.
  • On macOS, open the Anaconda Navigator and start Jupyter Notebooks from there.

Depending on your settings, you may get a new browser that points to your localhost or you may get a window with a web address that you need to copy and paste into a web browser (in which case, do that). In either case, the end result should be a web page open to the jupyter page with tabs for Files, Running, and Clusters.

Tutorial

There's a great tutorial at https://www.dataquest.io/blog/jupyter-notebook-tutorial/! You do not need to sign in or click the Start Free button to follow the tutorial.

A few notes:

• What is a Jupyter Notebook?
  • Nothing to add
• How to Follow This Tutorial
  • Nothing to add
• Installation
  • You can skip this if you already have Anaconda and have already started Jupyter Notebook in a browser.
• Creating Your First Notebook
  • You should already be at the Running Jupyter phase.
  • The New-> Python 3 might look like New->Python 3 (ipykernel)
  • In the Cells part:
    • CTRL-Enter or the $$\blacktriangleright\!\shortmid$$ runs the current cell; SHIFT-Enter runs the current cell and provides a new empty cell below it. SHIFT-Enter is generally the way to go as it runs the cell and gives you a new input line (rather than having to insert a new one)
    • ESC and ENTER toggle between command mode and edit mode. In edit mode, there is a pencil icon at the top right; in command mode, there isn't. Also, if you click in the edit part of a cell you enter edit mode; if you click in the space between the In []: and the >| you enter command mode.
  • In the Markdown part:
    • In addition to the commands shown, Markdown understands basic LaTeX (Greek letters, fractions, integrals, etc). Use single $ around commands for inline and $$ around commands for displaymath.
  • In the Kernels part:
    • For the commands that print formatted strings, the tutorial uses the string modulo method. To relate this to using format, and also to using the new (as of Python 3.6) f-string, here are three ways of printing the same information:

      a = 2
      b = 4
      c = 2.5
      d = 6.25
      # string modulo
      print('%d squared is %d and %0.2e squared is %0.2e' % (a, b, c, d))
      # format
      print('{:d} squared is {:d} and {:0.2e} squared is {:0.2e}'.format(a, b, c, d))
      # f-string
      print(f'{a:d} squared is {b:d} and {c:0.2e} squared is {d:0.2e}')
      

      You can see that all three are similar; the f-string puts the variable at the same location it will end up printing in the string rather than way at the end.

• Example Analysis
  • The "Setup" section in the middle of the page starts to go into some advanced data analysis with Pandas; they always give you the code, but it may be confusing! Also:
    • You will need to have saved their data file to the folder where you are saving your notebook. The file is in the "Example Data Analysis in a Jupyter Notebook" section way at the top of the page, or you can get it from https://s3.amazonaws.com/dq-blog-files/fortune500.csv.
    • There needs to be a carriage return after "import seaborn as sns"