Gapminder visualization

Hans Rosling

Most people do not understand the current world situation

Chimpanzee test

• 5 pairs of countries, one Asian, one European
• Which country has twice the child mortality of the other

Ploting

Panda plot
MatlibPlot
Seaborn

Plotly

Panda Plot

Wrapper around MatlibPlot

import pandas as pd
sample = pd.Series([1,2,4,8,16])
sample.plot()

<AxesSubplot:>

sample_df = pd.DataFrame( [[1,2,4,8,16],[1,3,9,27,71]])
sample_df.plot()

<AxesSubplot:>

sample_df.T.plot.line()

<AxesSubplot:>

sample_df = pd.DataFrame( {'two': [1,2,4,8,16],
                           'three': [1,3,9,27,71]})
sample_df.plot.line()

<AxesSubplot:>

sample_df.plot()

<AxesSubplot:>

sample_df.head()

	two	three
0	1	1
1	2	3
2	4	9
3	8	27
4	16	71

DataFrame Plot Methods

DataFrame.plot([x, y, kind, ax, ….])
DataFrame.plot.area([x, y])
DataFrame.plot.bar([x, y])
DataFrame.plot.barh([x, y])
DataFrame.plot.box([by])
DataFrame.plot.density([bw_method, ind])
DataFrame.plot.hexbin(x, y[, C, …])
DataFrame.plot.hist([by, bins])
DataFrame.plot.kde([bw_method, ind])
DataFrame.plot.line([x, y])
DataFrame.plot.pie([y])
DataFrame.plot.scatter(x, y[, s, c])
DataFrame.boxplot([column, by, ax, …])
DataFrame.hist([column, by, grid, …])

Docs

MatlibPlot

State machine

• Build up state about plot
• Then plot
• Jupyter clears state after running cell

Function and Object interface

import pandas as pd
import numpy as np 
from matplotlib import pyplot as plt


figure = plt.figure(figsize=(20,6))       # 20 inches by 6 inches?
axes1 = figure.add_subplot(2, 2, 1)       #add_subplot(nrows, ncols, index, **kwargs)
axes2 = figure.add_subplot(2, 2, 2)
axes3 = figure.add_subplot(2, 2, 3)

_ = axes1.hist(np.random.randn(100), bins=20, color='k', alpha=0.3)

axes2.scatter(np.arange(30), np.arange(30) + 3 * np.random.randn(30))

plt.plot(np.random.randn(50).cumsum(), 'k--')

[<matplotlib.lines.Line2D at 0x7fe1e96528e0>]

figure = plt.figure(figsize=(20,6),dpi=500)
plt.plot(np.random.randn(50).cumsum(), 'k--')

[<matplotlib.lines.Line2D at 0x7fe1e96c2a00>]

Terms

• Figures
• Axes (Plot)

    Main entry point for object interface

• Axis (x,y,z)
• Artist

    Everything you see

x = np.arange(0, 10, 0.2)
y = np.sin(x) 
figure, axes = plt.subplots(figsize=(20,6))
axes.plot(x, y)
plt.show()

matplotlib.pyplot.subplots(nrows=1, ncols=1, sharex=False, sharey=False, squeeze=True, subplot_kw=None, gridspec_kw=None, **fig_kw)

x = np.arange(0, 10, 0.2)
y = np.sin(x) 
figure = plt.figure(figsize=(20,6))
axes = figure.subplots()
axes.plot(x, y)
plt.show()

Saving plots to file

savefig saves the current active figure to a file

x = np.arange(0, 10, 0.2)
y = np.sin(x) 
figure, axes = plt.subplots(figsize=(20,6))
axes.plot(x, y)
plt.savefig('sample.png', dpi=400)

Matplotlib.pyplot Functions

Read the Docs

Matplotlib.pyplot.plot()

All those arguments

Seaborn

Examples from tutorial

Documentation

import pandas as pd
from matplotlib import pyplot as plt
import seaborn as sns

pokemon_df = pd.read_csv('Pokemon.csv', index_col=0)
pokemon_df.head()

	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

sns.lmplot(x='Attack', y='Defense', data=pokemon_df)

<seaborn.axisgrid.FacetGrid at 0x7fe20074c940>

seaborn.lmplot( x, y, data, hue=None, col=None, row=None,

    palette=None, col_wrap=None, height=5, aspect=1,

    markers='o', sharex=True, sharey=True, hue_order=None, 

    col_order=None, row_order=None, legend=True, legend_out=True,

    x_estimator=None, x_bins=None, x_ci='ci', scatter=True, fit_reg=True, 

    ci=95, n_boot=1000, units=None, order=1, logistic=False, lowess=False,

    robust=False, logx=False, x_partial=None, y_partial=None, truncate=False,

    x_jitter=None, y_jitter=None, scatter_kws=None, line_kws=None, size=None)

sns.lmplot(x='Attack', y='Defense', data=pokemon_df, height=7)

<seaborn.axisgrid.FacetGrid at 0x7fe218300070>

sns.lmplot(x='Attack', y='Defense', data=pokemon_df, aspect=0.5, height=7 )

<seaborn.axisgrid.FacetGrid at 0x7fe1e02f3280>

For Future Reference

Setting figsize does not work

plt.figure(figsize=(20,6)) 
sns.lmplot(x='Attack', y='Defense', data=pokemon_df)

<seaborn.axisgrid.FacetGrid at 0x7fe20a188f40>

<Figure size 1440x432 with 0 Axes>

plt.figure(figsize=(20,6)) 
sns.regplot(x='Attack', y='Defense', data=pokemon_df)

<AxesSubplot:xlabel='Attack', ylabel='Defense'>

plt.figure(figsize=(20,6)) 
sns.residplot(x='Attack', y='Defense', data=pokemon_df)

<AxesSubplot:xlabel='Attack', ylabel='Defense'>

sns.lmplot(x='Attack', y='Defense', data=pokemon_df,
           fit_reg=False, # No regression line
           hue='Stage')   # Color by evolution stage

<seaborn.axisgrid.FacetGrid at 0x7fe20a1bcd60>

sns.lmplot(x='Attack', y='Defense', data=pokemon_df,
           fit_reg=False, 
           hue='Stage')
 
# Tweak using Matplotlib
plt.ylim(0, None)
plt.xlim(0, None)

(0.0, 140.45)

Use Panda's to processs the data before ploting

pokemon_df = pd.read_csv('Pokemon.csv', index_col=0)
plt.figure(figsize=(20,6))                           # x axis = 20, y axis = 6
sns.boxplot(data= pokemon_df)

<AxesSubplot:>

plt.figure(figsize=(20,6))
stats_df = pokemon_df.drop(['Total', 'Stage', 'Legendary'], axis=1)
 
# New boxplot using stats_df
sns.boxplot(data=stats_df)

<AxesSubplot:>

Why repeat plt.figure(figsize=(20,6))

Jupyter resets plot settings at the end of each cell

Except for some settings

pairplot

Plot pairwise relationships in a dataset

pokemon_df = pd.read_csv('Pokemon.csv', index_col=0)
sns.pairplot(data=pokemon_df, vars=['Attack', 'Defense', 'Speed'], diag_kind='kde', plot_kws={'alpha': 0.2})

<seaborn.axisgrid.PairGrid at 0x7fe1d87babb0>

Seaside Plot Types

• Relational
• Categorical
• Distribution
• Regression
• Matrix

List of plot types

Themes

"white", "dark", "whitegrid", "darkgrid", "ticks"

plt.figure(figsize=(20,6))
sns.set_style('darkgrid')
 
# Violin plot
sns.violinplot(x='Type 1', y='Attack', data=pokemon_df)

<AxesSubplot:xlabel='Type 1', ylabel='Attack'>

Error Message

/anaconda/lib/python3.6/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use arr[tuple(seq)] instead of arr[seq]. In the future this will be interpreted as an array index, arr[np.array(seq)], which will result either in an error or a different result. return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval

Due to using python 3.7 or greater with scipy 1.1 or less

Color Palettes

pkmn_type_colors = ['#78C850',  # Grass
                    '#F08030',  # Fire
                    '#6890F0',  # Water
                    '#A8B820',  # Bug
                    '#A8A878',  # Normal
                    '#A040A0',  # Poison
                    '#F8D030',  # Electric
                    '#E0C068',  # Ground
                    '#EE99AC',  # Fairy
                    '#C03028',  # Fighting
                    '#F85888',  # Psychic
                    '#B8A038',  # Rock
                    '#705898',  # Ghost
                    '#98D8D8',  # Ice
                    '#7038F8',  # Dragon
                   ]

plt.figure(figsize=(20,6))
sns.violinplot(x='Type 1', y='Attack', data=pokemon_df, 
               palette=pkmn_type_colors)

<AxesSubplot:xlabel='Type 1', ylabel='Attack'>

Overlaying Plots

# First plot
plt.figure(figsize=(20,6))
sns.swarmplot(x='Type 1', y='Attack', data=pokemon_df, 
              palette=pkmn_type_colors)

<AxesSubplot:xlabel='Type 1', ylabel='Attack'>

plt.figure(figsize=(20,6))
sns.violinplot(x='Type 1', y='Attack', data=pokemon_df, 
               palette=pkmn_type_colors)

<AxesSubplot:xlabel='Type 1', ylabel='Attack'>

plt.figure(figsize=(20,6))
 
# Create plot
sns.violinplot(x='Type 1',
               y='Attack', 
               data=pokemon_df, 
               inner=None, # Remove the bars inside the violins
               palette=pkmn_type_colors)
 
sns.swarmplot(x='Type 1', 
              y='Attack', 
              data=pokemon_df, 
              color='k', # Make points black
              alpha=0.7) # and slightly transparent
 
# Set title with matplotlib
plt.title('Attack by Type')

Text(0.5, 1.0, 'Attack by Type')

pokemon_df = pd.read_csv('Pokemon.csv', index_col=0)
stats_df = pokemon_df.drop(['Total', 'Stage', 'Legendary'], axis=1)
corr = stats_df.corr()
 
plt.figure(figsize=(20,6))
# Heatmap
sns.heatmap(corr)

<AxesSubplot:>

plt.figure(figsize=(20,6))
sns.distplot(pokemon_df.Attack)

/Users/rwhitney/opt/anaconda3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

<AxesSubplot:xlabel='Attack', ylabel='Density'>

plt.figure(figsize=(20,6))
sns.countplot(x='Type 1', data=pokemon_df, palette=pkmn_type_colors)
 
# Rotate x-labels
plt.xticks(rotation=-45)

(array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
 [Text(0, 0, 'Grass'),
  Text(1, 0, 'Fire'),
  Text(2, 0, 'Water'),
  Text(3, 0, 'Bug'),
  Text(4, 0, 'Normal'),
  Text(5, 0, 'Poison'),
  Text(6, 0, 'Electric'),
  Text(7, 0, 'Ground'),
  Text(8, 0, 'Fairy'),
  Text(9, 0, 'Fighting'),
  Text(10, 0, 'Psychic'),
  Text(11, 0, 'Rock'),
  Text(12, 0, 'Ghost'),
  Text(13, 0, 'Ice'),
  Text(14, 0, 'Dragon')])

sns.jointplot(x='Attack', y='Defense', data=pokemon_df)

<seaborn.axisgrid.JointGrid at 0x7fe1e0678b80>

mpg = sns.load_dataset("mpg")
mpg.tail()

	mpg	cylinders	displacement	horsepower	weight	acceleration	model_year	origin	name
393	27.0	4	140.0	86.0	2790	15.6	82	usa	ford mustang gl
394	44.0	4	97.0	52.0	2130	24.6	82	europe	vw pickup
395	32.0	4	135.0	84.0	2295	11.6	82	usa	dodge rampage
396	28.0	4	120.0	79.0	2625	18.6	82	usa	ford ranger
397	31.0	4	119.0	82.0	2720	19.4	82	usa	chevy s-10

sns.lmplot(x='mpg', y='horsepower', data=mpg,fit_reg=False, hue='origin')

<seaborn.axisgrid.FacetGrid at 0x7fe20ad62670>

sns.lmplot(x='mpg', y='horsepower', data=mpg,fit_reg=False, hue='origin', height=6)

<seaborn.axisgrid.FacetGrid at 0x7fe20ad698b0>

sns.lmplot(x='mpg', y='weight', data=mpg, fit_reg=False,
           hue='origin', height=6)

<seaborn.axisgrid.FacetGrid at 0x7fe20ac46fa0>

sns.lmplot(y='mpg', x='model_year', data=mpg,
           hue='origin', height=6)

<seaborn.axisgrid.FacetGrid at 0x7fe1ea589f70>

plt.figure(figsize=(20,6))
#stats_df = mpg.drop(['Total', 'Stage', 'Legendary'], axis=1)
 
# New boxplot using stats_df
sns.boxplot(data=mpg, x='origin', y ='mpg')

<AxesSubplot:xlabel='origin', ylabel='mpg'>

mpg2019_df = pd.read_csv('2019mpg.csv', index_col=0)
mpg2019_df.head()

	MfrName	Division	Carline	Verify Mfr Cd	Index (Model Type Index)	Eng Displ	# Cyl	Transmission	City FE (Guide) - Conventional Fuel	Hwy FE (Guide) - Conventional Fuel	...	120V Charge time at 120 Volts (hours)	PHEV Total Driving Range (rounded to nearest 10 miles)DISTANCE	City PHEV Composite MPGe	Hwy PHEV Composite MPGe	Comb PHEV Composite MPGe	Unnamed: 162	Unnamed: 163	Unnamed: 164	Unnamed: 165	Unnamed: 166
Model Year
2019.0	Honda	Acura	NSX	HNX	41.0	3.5	6.0	Auto(AM-S9)	21.0	22.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2019.0	FCA US LLC	ALFA ROMEO	4C	CRX	403.0	1.8	4.0	Auto(AM6)	24.0	34.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2019.0	aston martin	Aston Martin Lagonda Ltd	Vantage V8	ASX	3.0	4.0	8.0	Auto(S8)	18.0	25.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2019.0	Volkswagen Group of	Audi	TT Roadster quattro	VGA	51.0	2.0	4.0	Auto(AM-S7)	23.0	31.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2019.0	BMW	BMW	Z4 sDrive30i	BMX	350.0	2.0	4.0	Auto(S8)	25.0	32.0	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

5 rows × 166 columns

plt.figure(figsize=(35,10))
sns.set(style="whitegrid")
sns.boxplot(data=mpg2019_df, x='MfrName', y ='mpg')

<AxesSubplot:xlabel='MfrName', ylabel='mpg'>