Index/ Python Data Analysis

---

Visualizations in Python

Objective

Learn graphing and visualization in Python

Reference

Matplotlib is the workhorse of python visualization. It is old but widely used.

Seaborn library aims to give pretty graphs and high level APIs.

Essentials (★☆☆)

Generic

• 5 Data Visualization Best Practices

Visualization in Python

• Graph libraries in Python - A good overview of graphics libraries in Python
• Python graph gallery is the ultimate website with lots of sample graph and code for both Matplot and Seaborn
• Visualization section from the excellent open source book Python Data Science Book
• Seaborn aims to give prettier graphs out of the box
  • Seaborn section from the excellent open source book Python Data Science Book
  • An excellent Seaborn tutorial with Pokemon dataset :-)

Checklist

After completing the exercises below, you should be comfortable with

• Creating visualizations using matplot and seaborn
• What is the difference between Seaborn vs Matplotlib?

Exercises

Difficulty Level

★☆☆ - Easy
★★☆ - Medium
★★★ - Challenging
★★★★ - Bonus

A - Easy

A1 - Do a scatter plot of following data (★☆☆)

bills = [50,30,60,40,65,20,10,15,25,35]
tips= [12,7,13,8,15,5,2,2,3,4]

• Label x-axis as ‘bill’
• Label y-axis as ‘tip’

A2 - Do a boxplot of the following data (★☆☆)

a = [22, 25, 30, 35, 40, 42, 45, 50, 55, 60, 65, 70]

A3 - Do a lineplot of following data (★☆☆)

month  revenue
Jan    10
Feb    12
Mar     7
Apr    15
May    17

A4 - Do a barplot of above data (★☆☆)

A5 - Do a pie chart of following data (★☆☆)

Gender  percentage
M        52
F        40
Unknown   8

A6 - Plot a histogram of data (★☆☆)

Exercises

EX-1: Visualize Housing Dataset

• Data: house-sales.csv
• Visualize the following
  • Number of sales per bedrooms (bar plot or Pie)
  • Average house price per bedrooms (box plot)
  • Relationship between ‘SQFT’ vs ‘Sale Price’ (Scatter plot)
  • Any other interesting plots you can come up with

EX-2 - Visualize pokemon data (★★☆)

Download pokemon data

Read it like this:

pokemon = pd.read_csv('https://s3.amazonaws.com/elephantscale-public/data/pokemon/pokemon-small.csv', index_col=0)

with pd.option_context("display.width", 150):
    print (pokemon)

Data looks like this:

           Name   Type 1  Type 2  Total   HP  Attack  Defense  Sp. Atk  Sp. Def  Speed  Stage  Legendary
#                                                                                                       
1     Bulbasaur    Grass  Poison    318   45      49       49       65       65     45      1      False
2       Ivysaur    Grass  Poison    405   60      62       63       80       80     60      2      False
3      Venusaur    Grass  Poison    525   80      82       83      100      100     80      3      False
4    Charmander     Fire     NaN    309   39      52       43       60       50     65      1      False
5    Charmeleon     Fire     NaN    405   58      64       58       80       65     80      2      False
..          ...      ...     ...    ...  ...     ...      ...      ...      ...    ...    ...        ...
147     Dratini   Dragon     NaN    300   41      64       45       50       50     50      1      False
148   Dragonair   Dragon     NaN    420   61      84       65       70       70     70      2      False
149   Dragonite   Dragon  Flying    600   91     134       95      100      100     80      3      False
150      Mewtwo  Psychic     NaN    680  106     110       90      154       90    130      1       True
151         Mew  Psychic     NaN    600  100     100      100      100      100    100      1      False

Try the following graphs:

EX-1A - Do a graph to illustrate how many type-1 Pokemons
Hint: histogram

EX-1B - Illustrate Attack points per Type-1 in a boxplot
Hint: You may need to find the average attack points per type first

Ex-1C - Come up with a graph to explain this data. Experiment

More Exercices

---

Index/ Python Data Analysis