Python crash

course
 Overview

Why Python

The plan Install Anaconda

Jupyter Notebook walkthrough

Q&A and support

2
 Why Python
• Python is a top tool for data science
• Many DS jobs require Python skills
• Many DS tools built in Python
• Network effects

3
 Python Philosiphy
Python is meant to be fun, simple, clear, and easy to use.
There are some parts of “The Zen of Python” that describe
this:
• Beautiful is better than ugly.
• Explicit is better than implicit.
• Simple is better than complex.
• Complex is better than complicated.
• Readability counts.
• "there should be one— and preferably only one —
obvious way to do it"

4
Network effects
• The more users, the bigger the community and more
support available. More tools are built.
• Julia is a newer language and may be more viable as a
top data science language in 5-10 years.
• Many other tools out there to do data science with,
including GUIs such as RapidMiner and Excel.

https://www.kdnuggets.com/2018/05/poll-tools-analytics-data- 5
science-machine-learning-results.html
 Python distributions and installations
• Python distributions are one easy way to install Python, a • I also use Windows Terminal, although Anaconda Prompt
package manager, and some Python packages and Powershell work too. To use Terminal or another
Powershell we should do ‘conda init powershell’ from
• Anaconda is what we’ll use, although there are many Anaconda Powershell.
others like Python(x,y) and more.
• On Mac or Linux we can use Terminal for command line
• Download Anaconda and install. access.
• On Windows, I like to check the box ‘add to path’ even • We can open a Jupyter Notebook from Anaconda
though it’s not recommended. Navigator or through a Terminal or command prompt by
typing ‘jupyter notebook’.
• We can also use IPython (ipython from the command
line) or Jupyter Lab to run Python code.
• There is also a Python shell (by typing ‘python’ from a
terminal) but it is limited.

6
 Resources for the program and
Python/R
• Your professor/instructor is here to help
• Your advisor and success coach
• Other students can help too (e.g. via discussions)
• Regis Learning Commons
• Writing Center and Tutoring
• Documentation (e.g. docs.python.org)
• Stack Overflow
• Search engines (google, duckduckgo)
• Archives for broken links (archive.is, archive.org)
Python benevolent dictator for
life (Resigned in July 2018)

7
 Docs and books
Using a search engine and stackoverflow is part of writing
code and doing data science!
Books (e.g. through the library and O’Reilly Safari) are a
good way to understand fundamentals of statistics, machine
learning, and specialized topics.
ISLR (stats in R)
ESL (stats methods and math)
Python Machine Learning - Third Edition (Packt)
Clean code in Python – Second Edition (Packt)
R for Data Science

8
 Extra Resources for Learning Python
Python is a foundational skill for much of modern data science. It is
worth putting in some time to build solid Python skills.

If you are still learning Python or feeling shaky on the basics If you are feeling relatively comfortable with Python, you
(even after doing the readings), consider doing the following: might consider doing the Kaggle course on Python if you
need a refresher: https://www.kaggle.com/learn/python
Complete most of the Udacity Intro to Python Course.
• Weeks 1 and 2: Lessons 1 and 2
There are many other Python learning resources for all
• Week 3 and 4: Lesson 3
levels listed here:
• Week 5 and 6: Lesson 4 https://forums.fast.ai/t/recommended-python-learning-resour
ces/26888
• Week 7 and 8: Lesson 5

9
Ways to continue building your
Python and R skills
• DataCamp
• DataQuest
• Hackerrank
• Reading the official Python documentation or
documentation for other Python packages
• Kaggle (participating in competitions, looking at what
others are doing)
• Codewars.com
• Books (e.g. Clean Code with Python 2nd Edition by
Mariano Anaya), Minimal Python by Noah Gift and
Alfredo Deza, Python Data Science Essentials by Alberto
Boschetti, etc)

10