Density Transform#

The density transform performs one-dimensional kernel density estimation over input data and generates a new column of samples of the estimated densities.

Here is a simple example, showing the distribution of IMDB ratings from the movies dataset:

import altair as alt
from vega_datasets import data

alt.Chart(data.movies.url).transform_density(
    'IMDB_Rating',
    as_=['IMDB_Rating', 'density'],
).mark_area().encode(
    x="IMDB_Rating:Q",
    y='density:Q',
)

The density can also be computed on a per-group basis, by specifying the groupby argument. Here we split the above density computation across movie genres:

import altair as alt
from vega_datasets import data

alt.Chart(
    data.movies.url,
    width=120,
    height=80
).transform_filter(
    'isValid(datum.Major_Genre)'
).transform_density(
    'IMDB_Rating',
    groupby=['Major_Genre'],
    as_=['IMDB_Rating', 'density'],
    extent=[1, 10],
).mark_area().encode(
    x="IMDB_Rating:Q",
    y='density:Q',
).facet(
    'Major_Genre:N',
    columns=4
)

Transform Options#

The transform_density() method is built on the DensityTransform class, which has the following options:

Property

Type

Description

as

array(FieldName)

The output fields for the sample value and corresponding density estimate.

Default value: ["value", "density"]

bandwidth

number

The bandwidth (standard deviation) of the Gaussian kernel. If unspecified or set to zero, the bandwidth value is automatically estimated from the input data using Scott’s rule.

counts

boolean

A boolean flag indicating if the output values should be probability estimates (false) or smoothed counts (true).

Default value: false

cumulative

boolean

A boolean flag indicating whether to produce density estimates (false) or cumulative density estimates (true).

Default value: false

density

FieldName

The data field for which to perform density estimation.

extent

array(number)

A [min, max] domain from which to sample the distribution. If unspecified, the extent will be determined by the observed minimum and maximum values of the density value field.

groupby

array(FieldName)

The data fields to group by. If not specified, a single group containing all data objects will be used.

maxsteps

number

The maximum number of samples to take along the extent domain for plotting the density.

Default value: 200

minsteps

number

The minimum number of samples to take along the extent domain for plotting the density.

Default value: 25

steps

number

The exact number of samples to take along the extent domain for plotting the density. If specified, overrides both minsteps and maxsteps to set an exact number of uniform samples. Potentially useful in conjunction with a fixed extent to ensure consistent sample points for stacked densities.