This notebook was created by Jean de Dieu Nyandwi for the love of machine learning community. For any feedback, errors or suggestion, he can be reached on email (johnjw7084 at gmail dot com), Twitter, or LinkedIn.

Data Visualization with Seaborn¶

Seaborn is a fantastic and easy to use Python Visualization which is built on Matplotlib.

For a quick look, check out the gallery page.

To be covered:

• 1. Relational Plots:
  • Scatter plots
  • Line plots

• 2. Distribution Plots:

  • Plotting Histograms with displot() and histplot()
  • Plotting Bivariate Data with Jointplot()
  • Plotting Many Distribution with pairplot()
  • Plotting Distributions with rugplot()
  • Kernel Density Estimation (KDE) Plot with kdeplot() and displot()
  • Cumulative Distributions

• 3. Categorical Plots:

  • Categorical estimate plots
  • Categorical distribution plots
  • Categorical scatter plots
  • Plotting multiple categorical plots

• 4. Regression Plots

• 5. Multiplots:

  • Facet grids
  • Pair grids
  • Pair plots

• 6. Matrix Plots: Heat and Cluster Maps

  • Heat maps
  • Cluster maps

• 7. Style and Color

  • Styles and themes
  • Colors

In this lab, we will use real world datasets, which are already part of Seaborn.

1. Relational Plots¶

These kind of plots are used to analyze the relationship between features.

• Scatter Plots
• Line Plots

Imports¶

In [166]:

import seaborn as sns
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

import seaborn as sns import numpy as np import pandas as pd import matplotlib.pyplot as plt

In [167]:

# Loading the datasets to be used in this lab

titanic = sns.load_dataset('titanic')
fmri = sns.load_dataset('fmri')
tips = sns.load_dataset('tips')
flights = sns.load_dataset('flights')

# Loading the datasets to be used in this lab titanic = sns.load_dataset('titanic') fmri = sns.load_dataset('fmri') tips = sns.load_dataset('tips') flights = sns.load_dataset('flights')

In [168]:

titanic.head()

titanic.head()

Out[168]:

	survived	pclass	sex	age	sibsp	parch	fare	embarked	class	who	adult_male	deck	embark_town	alive	alone
0	0	3	male	22.0	1	0	7.2500	S	Third	man	True	NaN	Southampton	no	False
1	1	1	female	38.0	1	0	71.2833	C	First	woman	False	C	Cherbourg	yes	False
2	1	3	female	26.0	0	0	7.9250	S	Third	woman	False	NaN	Southampton	yes	True
3	1	1	female	35.0	1	0	53.1000	S	First	woman	False	C	Southampton	yes	False
4	0	3	male	35.0	0	0	8.0500	S	Third	man	True	NaN	Southampton	no	True

In [169]:

fmri.head(3)

fmri.head(3)

Out[169]:

	subject	timepoint	event	region	signal
0	s13	18	stim	parietal	-0.017552
1	s5	14	stim	parietal	-0.080883
2	s12	18	stim	parietal	-0.081033

In [170]:

tips.head(3)

tips.head(3)

Out[170]:

	total_bill	tip	sex	smoker	day	time	size
0	16.99	1.01	Female	No	Sun	Dinner	2
1	10.34	1.66	Male	No	Sun	Dinner	3
2	21.01	3.50	Male	No	Sun	Dinner	3

In [171]:

flights.head(3)

flights.head(3)

Out[171]:

	year	month	passengers
0	1949	Jan	112
1	1949	Feb	118
2	1949	Mar	132

Scatter Plots¶

In order to visualize the relationship between between two numeric features, scatter plot can be a go to plot over other types.

We will use sns.scatterplot(data, x, y, hue, style, palette, size, sizes, legend, markers...) and also sns.relplot().

In [172]:

sns.scatterplot(data=titanic, x='age', y='fare')

sns.scatterplot(data=titanic, x='age', y='fare')

Out[172]:

<AxesSubplot:xlabel='age', ylabel='fare'>

With hue parameter, we can map another feature to the plot.

In [173]:

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex')

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex')

Out[173]:

<AxesSubplot:xlabel='age', ylabel='fare'>

You can see it makes the plots more clear. In terms of this titanic dataset, you can directly identify that women paid high fare than men.

To also highlight the difference between the hue classes, we can add marker style as follows.

In [174]:

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='sex')

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='sex')

Out[174]:

<AxesSubplot:xlabel='age', ylabel='fare'>

Increasing the figure size...

In [175]:

plt.figure(figsize=(8,6))
sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='pclass')

plt.figure(figsize=(8,6)) sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='pclass')

Out[175]:

<AxesSubplot:xlabel='age', ylabel='fare'>

You can also use specific markers with style and markers paremeters.

In [176]:

plt.figure(figsize=(8,6))

markers = {1:'P', 2:'X', 3:'D'} # P, X, and D are markers
sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='pclass', markers=markers)

plt.figure(figsize=(8,6)) markers = {1:'P', 2:'X', 3:'D'} # P, X, and D are markers sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', style='pclass', markers=markers)

Out[176]:

<AxesSubplot:xlabel='age', ylabel='fare'>

Adding same feature on hue to size will make the plot more meaninfgul. sizes will control the range of marker areas of size. The

In [177]:

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', size='sex', sizes=(20,200))

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', size='sex', sizes=(20,200))

Out[177]:

<AxesSubplot:xlabel='age', ylabel='fare'>

In [178]:

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', size='pclass', sizes=(20,200))

sns.scatterplot(data=titanic, x='age', y='fare', hue='sex', size='pclass', sizes=(20,200))

Out[178]:

<AxesSubplot:xlabel='age', ylabel='fare'>