Interactive Scatter Plot and Linked Layered Histogram

This example shows how to link a scatter plot and a histogram together such that clicking on a point in the scatter plot will isolate the distribution corresponding to that point, and vice versa.

Method syntax

import altair as alt
import pandas as pd
import numpy as np

# generate fake data
source = pd.DataFrame({
    'gender': ['M']*1000 + ['F']*1000,
    'height':np.concatenate((
        np.random.normal(69, 7, 1000), np.random.normal(64, 6, 1000)
    )),
    'weight': np.concatenate((
        np.random.normal(195.8, 144, 1000), np.random.normal(167, 100, 1000)
    )),
    'age': np.concatenate((
        np.random.normal(45, 8, 1000), np.random.normal(51, 6, 1000)
        ))
    })

selector = alt.selection_point(fields=['gender'])

color_scale = alt.Scale(domain=['M', 'F'],
                        range=['#1FC3AA', '#8624F5'])
color = (
    alt.when(selector)
    .then(alt.Color("gender:N").scale(color_scale))
    .otherwise(alt.value("lightgray"))
)
base = alt.Chart(source).properties(
    width=250,
    height=250
).add_params(selector)

points = base.mark_point(filled=True, size=200).encode(
    alt.X('mean(height):Q').scale(domain=[0,84]),
    alt.Y('mean(weight):Q').scale(domain=[0,250]),
    color=color,
)

hists = base.mark_bar(opacity=0.5, thickness=100).encode(
    alt.X('age')
        .bin(step=5) # step keeps bin size the same
        .scale(domain=[0,100]),
    alt.Y('count()')
        .stack(None)
        .scale(domain=[0,350]),
    alt.Color('gender:N').scale(color_scale)
).transform_filter(
    selector
)

points | hists

Attribute syntax

import altair as alt
import pandas as pd
import numpy as np

# generate fake data
source = pd.DataFrame({
    'gender': ['M']*1000 + ['F']*1000,
    'height':np.concatenate((
        np.random.normal(69, 7, 1000), np.random.normal(64, 6, 1000)
    )),
    'weight': np.concatenate((
        np.random.normal(195.8, 144, 1000), np.random.normal(167, 100, 1000)
    )),
    'age': np.concatenate((
        np.random.normal(45, 8, 1000), np.random.normal(51, 6, 1000)
        ))
    })

selector = alt.selection_point(fields=['gender'])

color_scale = alt.Scale(domain=['M', 'F'],
                        range=['#1FC3AA', '#8624F5'])
color = (
    alt.when(selector)
    .then(alt.Color("gender:N", scale=color_scale))
    .otherwise(alt.value("lightgray"))
)
base = alt.Chart(source).properties(
    width=250,
    height=250
).add_params(selector)

points = base.mark_point(filled=True, size=200).encode(
    x=alt.X('mean(height):Q', scale=alt.Scale(domain=[0,84])),
    y=alt.Y('mean(weight):Q', scale=alt.Scale(domain=[0,250])),
    color=color,
)

hists = base.mark_bar(opacity=0.5, thickness=100).encode(
    x=alt.X('age',
            bin=alt.Bin(step=5), # step keeps bin size the same
            scale=alt.Scale(domain=[0,100])),
    y=alt.Y('count()',
            stack=None,
            scale=alt.Scale(domain=[0,350])),
    color=alt.Color('gender:N',
                    scale=color_scale)
).transform_filter(
    selector
)


points | hists