Dinosaurs

#30DayChartChallenge Day 19

Today, our theme is dinosaurs, which made me think about the Datasaurus Dozen! This wild set of datasets is an extension of the famous Anscombe’s Quartet. Despite having completely different-looking distributions when plotted, every single one of the Datasaurus Dozen datasets has the exact same summary statistics.

The Datasaurus Dozen is a great reminder for all us data folks to always look beyond summary statistics like the mean and standard deviation to truly understand what’s going on with your data.

import polars as pl
import plotly.express as px

df = pl.read_csv('./datasaurus.csv')

fig = px.scatter(df, 
    x='x', y='y', color='x',
    animation_frame='dataset', 
    range_x=[df['x'].min(), df['x'].max()], 
    range_y=[df['y'].min(), df['y'].max()],
    template='simple_white',
    width=500, height=500,
)

fig.layout.updatemenus[0].buttons[0].args[1]["frame"]["duration"] = 1000
fig.update_coloraxes(showscale=False)
fig.update_layout(
    title= 'The Datasaurus Dozen<br><sup>Each of these datasets share the same summary statistics</sup>',
    xaxis_title=None,
    yaxis_title=None,
    margin = dict(l=0, r=0, b=45),
    showlegend=False,
    annotations = [dict(
        x=1,
        y=-0.22, 
        xref='paper',
        yref='paper',
        text='Source: <a href="https://www.research.autodesk.com/publications/same-stats-different-graphs/">https://www.research.autodesk.com/publications/same-stats-different-graphs/</a><br>                                                                                   Made by: www.ddanieltan.com',
        showarrow = False,
        font = dict(size=10, color='grey')
    )]
)
fig

TIL

1. In general, I find Plotly’s configurations quite unintuitive and a lot of the answers that I look for end up in a forum/StackOverflow post somewhere online:

• [Configuration to slow down animation was found nested in this community post] (https://community.plotly.com/t/how-to-slow-down-animation-in-plotly-express/31309/5)

• Configuration to remove the colorbar was found in this Stack Overflow