Implementation:DistrictDataLabs Yellowbrick Datasaurus Plot

Knowledge Sources	DistrictDataLabs_Yellowbrick Same Stats Different Graphs
Domains	Visualization, Data_Science
Last Updated	2026-02-08 05:00 GMT

Overview

Concrete tool for rendering the Datasaurus Dozen demonstration plot illustrating why visualization is essential for data analysis, provided by the Yellowbrick library.

Description

The datasaurus module embeds four datasets from the Datasaurus Dozen (Matejka & Fitzmaurice, 2017 CHI Conference). These datasets have nearly identical summary statistics (mean, variance, correlation) yet produce dramatically different scatter plots. The module renders a 2x2 grid with scatter plots and linear best fit lines for each dataset.

Usage

Import this function for teaching or demonstration contexts where you want to illustrate the importance of data visualization beyond summary statistics.

Code Reference

Source Location

• Repository: DistrictDataLabs_Yellowbrick
• File: yellowbrick/datasaurus.py
• Lines: 1-1235

Signature

def datasaurus():
    """
    Creates 2x2 grid plot of 4 Datasaurus Dozen datasets.
    Each subplot shows scatter points and a linear best fit line.

    Returns
    -------
    axa, axb, axc, axd : tuple of matplotlib.Axes
    """

Import

from yellowbrick.datasaurus import datasaurus

I/O Contract

Inputs

Name	Type	Required	Description
(none)	—	—	Function takes no arguments

Outputs

Name	Type	Description
axes	tuple of 4 Axes	The four matplotlib Axes objects (axa, axb, axc, axd)

Usage Examples

from yellowbrick.datasaurus import datasaurus
import matplotlib.pyplot as plt

# Display the Datasaurus Dozen demo
axes = datasaurus()
plt.tight_layout()
plt.show()

Related Pages

• Environment:DistrictDataLabs_Yellowbrick_Python_Scikit_Learn_Environment