Make data visualization with Seaborn

Why seaborn?

In previous machine learning examples, we have already seen some nice plots and visulizaions which will definetely help us. In this exercise, we could practice some other visualization techniques.

This time we will still play with Iris data. Please get familar with that data first. In the exercise, you could learn how to make plots such as bloxplot, pairplot and some other forms of data visualizations. If you don't know what does it exactly mean, please search material online.



In [8]:

    
# please watch out how we import seaborn package and how we rename it as sns
import seaborn as sns



In [6]:

    
import pandas as pd
# read CSV file directly from a URL and save the results
iris = pd.read_csv('https://raw.githubusercontent.com/uiuc-cse/data-fa14/gh-pages/data/iris.csv')



In [3]:

    
iris.head()









    Out[3]:






  
    
      
      sepal_length
      sepal_width
      petal_length
      petal_width
      species
    
  
  
    
      0
      5.1
      3.5
      1.4
      0.2
      setosa
    
    
      1
      4.9
      3.0
      1.4
      0.2
      setosa
    
    
      2
      4.7
      3.2
      1.3
      0.2
      setosa
    
    
      3
      4.6
      3.1
      1.5
      0.2
      setosa
    
    
      4
      5.0
      3.6
      1.4
      0.2
      setosa



In [4]:

    
iris["species"].value_counts()









    Out[4]:





versicolor    50
setosa        50
virginica     50
Name: species, dtype: int64

Joint Plot



In [7]:

    
# This is not a waste line, please search information why we need this.
%matplotlib inline



In [10]:

    
# This is a joint plot for two variables sepal_length and sepal_width
sns.jointplot(x ="sepal_length", y = "sepal_width", data =iris, size= 5)









    Out[10]:





<seaborn.axisgrid.JointGrid at 0x155ca6fc160>

Problem 1. Please make another joint plot for petal_length and petal_wdith



In [ ]:

    
# Your code and then you run it:



In [11]:

    
import matplotlib.pyplot as plt
sns.FacetGrid(iris, hue = "species", size =5)\
    .map(plt.scatter, "sepal_length", "sepal_width")\
    .add_legend()









    Out[11]:





<seaborn.axisgrid.FacetGrid at 0x155ca904f28>

Boxplot



In [12]:

    
ax = sns.boxplot(x = "species", y = "petal_length", data = iris)

Problem 2. Please make another box plot to compare petal width for different speicies.



In [ ]:

    
# Plese put your code here and then run it

Optional problem.

Please read http://seaborn.pydata.org/generated/seaborn.pairplot.html. Make a pairplot for all the variables and make the type as scatter.



In [ ]:

    
# Code here and run it.

	sepal_length	sepal_width	petal_length	petal_width	species
0	5.1	3.5	1.4	0.2	setosa
1	4.9	3.0	1.4	0.2	setosa
2	4.7	3.2	1.3	0.2	setosa
3	4.6	3.1	1.5	0.2	setosa
4	5.0	3.6	1.4	0.2	setosa