Lecture 23 – Introduction to Visualization

Data 6, Summer 2021

# Notice the new import statements today!

from datascience import *
import numpy as np
import seaborn as sns

Table.interactive_plots()

What is visualization?

# Run this cell.
full_titanic = Table.from_df(sns.load_dataset('titanic').dropna())
titanic = full_titanic.select('sex', 'age', 'fare')
titanic

sex	age	fare
female	38	71.2833
female	35	53.1
male	54	51.8625
female	4	16.7
female	58	26.55
male	34	13
male	28	35.5
male	19	263
female	49	76.7292
male	65	61.9792

... (172 rows omitted)

Visualizations are for humans!

titanic.select('age', 'fare').scatter('age', 'fare', 
                width = 500, 
                height = 500,
                title = 'Fare vs. age for Titanic passengers',
                show = False)

Visualize, then quantify!

ans = Table.from_df(sns.load_dataset('anscombe'))
ans

dataset	x	y
I	10	8.04
I	8	6.95
I	13	7.58
I	9	8.81
I	11	8.33
I	14	9.96
I	6	7.24
I	4	4.26
I	12	10.84
I	7	4.82

... (34 rows omitted)

import plotly.graph_objects as go
from plotly.subplots import make_subplots

fig = make_subplots(rows=2, cols=2, start_cell="bottom-left")

for ds, pos in zip(['I', 'II', 'III', 'IV'], [(2, 1), (2, 2), (1, 1), (1, 2)]):
    fig.add_trace(go.Scatter(
            x = ans.where('dataset', ds).column('x'), 
            y = ans.where('dataset', ds).column('y'), 
            mode = 'markers',
            name = 'Dataset ' + ds),
        row=pos[0], col=pos[1])

fig.update_layout(title = "Anscombe's Quartet")

fig.show()

Statistical variable types

bay_codes = Table().with_columns(
    'city', np.array(['Berkeley', 'San Francisco', 'Palo Alto']),
    'area code', np.array([510, 415, 650])
)

bay_codes

city	area code
Berkeley	510
San Francisco	415
Palo Alto	650

Encoding

titanic.group('sex', np.mean).select('sex', 'age mean').barh('sex', 'age mean', title = 'Average age of females and males on the Titanic', show = False)