Welcome to Foundations of Python

Hamish Gibbs

Course aims

  • A foundation in Python programming.
    • Variables, data structures, control logic, functions, classes.
  • An Introduction to popular Python tools for data science.
    • pandas, matplotlib, sklearn.
  • A hands-on data science challenge.
    • Predicting the price of AirBnBs in London.

Also: tools of ‘professional’ Python

  • Using an Integrated Development Environment
    • VSCode
  • Using collaboration tools
    • git, GitHub

Course format

  • This is a short course!
  • Days 1 – 3: lectures and practicals.
    • Practical sessions rely on existing Python tutorials.
    • Once you have worked through a practical, try to change the examples it provides until you “get it”.
    • We are using a variety of open source Python learning resources.
  • Day 4: Working together on a collaborative challenge.

Schedule

  • Day 1: Python basics
    • Variables, data structures (list, dict), control logic (if, for, while).
  • Day 2: Abstraction & composition
    • Functions (def), Classes (class).
    • Also: Using .py files, not .ipynb in VSCode.

Schedule

  • Day 3: Python data science
    • pandas, numpy, matplotlib.
    • Also: Collaboration with git and GitHub.
  • Day 4: Challenge: regression analysis
    • Predicting the price of London AirBnBs using Inside AirBnB data.

About you

  • Programming experience?
  • Statistics experience?
  • Any installation problems?

About me

  • Final year PhD student
    • Supervised by James Cheshire
  • My research interests:
    • Human mobility, disease transmission, bias & uncertainty
  • Python experience:
    • 9 years
Gibbs et. al., Detecting behavioural changes in human movement to inform the spatial scale of interventions against COVID-19, PLOS Computational Biology (2021)

Learning python

Source: Sarah’s Scribbles

Learning python

  • Practice is the most important ingredient to becoming a good programmer.
  • It is easier to “practice” if you find personally compelling reasons to use Python.
    • Coursework, side projects, random curiosity, automating things in your life.
  • Programming is all about trial and error.

AI

  • New AI programming assistants:
    • Chat GPT, GitHub Copilot, Copilot Chat.
  • I recommend using them all, especially as a study aid.
    • Bad idea: Using AI to generate code you can’t understand.
    • Good idea: Using AI to explain code you can’t understand.

Any questions?

Basic data types

Integer 

as.integer(10) # R
int(10) # Python

Float 

as.numeric(10.3) # R
float(10.3) # Python

String 

as.character("Hello") # R
str("Hello") # Python

Type checking 

class("Hello") # R
type("Hello") # Python

Tutorial #1: Variables

  • Variables, expressions, and statements
  • Core concepts:

    • Variable assignment and basic math

      minute = 20
minute + 32

    • Working with strings

      first = '100'
second = '150'
print(first + second)

    • Subsetting strings

      word = 'Python'
word[0]

Tutorial #2: Lists

  • An informal introduction to python §3.1.3
  • Core concepts:

    • List indexing

      squares = [1, 4, 9, 16, 25]
squares[0]

    • List manipulation

      squares + [36, 49, 64, 81, 100]
squares.append(10)

    • List mutability

      rgb = ["Red", "Green", "Blue"]
rgba = rgb
id(rgb) == id(rgba)  # they reference the same object

One more data structure: tuples

  • A tuple is an immutable collection of values.
coord = (0, 1) # a single tuple
coords = [coord, coord] # a list of tuples
  • Unlike a list, the values in a tuple cannot be changed.
    • This means no .append() or .sort() methods (which both change the values in a list).
  • Tuples are faster than lists and good when you have a collection of values that won’t have to change once it has been created.

Just one more data structure: set

  • A set is an unordered collection with no duplicate elements.

    set1 = {0, 1} # a set
set2 = set([0, 1]) # creating a set from a list

  • A set is mutable (its values can be changed).

  • It is very quick to check if an element is in a set:

    1 in {0, 1} # I'm faster!
1 in [0, 1]
    • This doesn’t matter for a collection of 2 values, but matters a lot for larger collections.

Tutorial #3: Dictionaries

  • Dictionaries §1
    • To start, only work on the first section of this tutorial!
    • If you “get it,” move on to this afternoon’s tutorials
  • Core concepts:

    • The {key: value} format of dictionaries

      eng2sp = dict()
eng2sp['one'] = 'uno'
eng2sp['one']

    • The .items(), .keys(), and .values() methods

Using Python

Setting up Colab

  • Does everyone have a Google account?

  • Go to colab.research.google.com.

  • Open a new notebook.

  • Rename your notebook.

  • Get started!