Files and code for a Python 1 hands-on session at NICAR18 in Chicago

Outline for hands-on session

1. Open a Jupyter notebook file (link to download)

   • Remember - hit Shift-Enter to run a cell!

2. Overview of why this is a good language to learn on, how it may be different from others

   • Reliance on indentation
   • Loose vs. strict typing
   • Lots of built in functions and modules that can be imported
   • (Show XKCD comic here: https://xkcd.com/353/)
   • Do basic math
   • Print out some text
   • Integrate result in a sentence
   • Print it out

3. Data types

   • Variables
   • Numbers
   • Strings
     • Important functions:
       • Slice (use numeric indices, -1 counts from end)
       • Lower/upper
       • Strip
       • Startswith
       • Find - look for a string within a string
   • Booleans
   • Lists
     • Join function
   • Dictionaries

4. Iteration

   • Demonstrate sample with a list like ["dog", "cat"]
   • Take our data file and load each row individually, to show what looping is

5. Comments

   • Why do we use comments?
   • How do we add them to a file?

6. Importing modules

   • Load a sample module
   • Where do we load it?
   • What name do we use when referring to it?
   • What's the difference between targeting a library and a module? (ex: from x import y)
   • import datetime oTime = datetime.datetime.now() print oTime.isoformat()

7. Pip (to gain access to more modules)

   • explain how some modules built in/some can be pulled down
   • walk through online directory of what modules exist
     • PyPi index
     • https://wiki.python.org/moin/UsefulModules#Database
   • pull one down - like pip install pandas

8. Start use of Pandas to load and manipulate file

   • find current directory with os.getcwd()
   • We will use this file: https://catalog.data.gov/dataset/sex-offenders-938b3
   • import pandas as pd
   • input = pd.read_csv(PATH_TO_SEX_OFFENDERS_DATABASE);
   • #row
   • input.iloc[1]
   • #column
   • input['TOTAL CASES']
   • #row and column
   • input.iloc[3]['TOTAL CASES']