Insight Data Science

Jupyter Notebook (formerly IPython Notebook) + Python + Data Science Workflow

By: Jonathan Whitmore

• Email: jbwhit@gmail.com
• Twitter: @jbwhitmore
• LinkedIn: Add me w/ 1 line message about Insight

Notebooks: https://github.com/jbwhit/jupyter-best-practices

Things to comment on about Insight

• 10 things
• You will be annoyed with meeting people who want to hire you.

Table of Contents

1. Initial Couple of Demos to demonstrate some of the Notebook's capabilities
2. Exploratory (Interactive) Data Analysis
3. Some python and coding basics
4. Notebook specific basics
5. Deeper in Pandas, Plotting and SQL
6. Extras if time!

Themes

• Workflow matters
  • Make repeated actions as automatic as possible
• Adhere to best practices as best you can (unless you have a really good reason not to).
  1. You will have to collaborate on teams -- code quality matters.
  2. PEP8 things
• Learn your tools
  • there are csv parsers, don't roll your own
• Contribute back -- stackoverflow

Resources

• https://github.com/ipython/ipython/wiki/A-gallery-of-interesting-IPython-Notebooks
• git and github
  • Creator of IPython, Fernando Perez's, git/github tutorial via IPython Notebook based on Insight Fellow John McDonnell.
  • Jake Vanderplas's tutorial
• Stitch Fix Post
• Peter Norvig's Notebook on Traveling Salesman

Final comment

• Don't try to take notes, just let this wash over you.
• These Notebooks will be available to you.