Themes and styling

Theme holds presentation choices; data and encodings stay on Plot. Calling .theme() attaches a complete theme, so construct the final theme before attaching it rather than expecting several theme calls to merge.

from dataviz import Plot, Theme, save
from dataviz.core.colors import SEAGREEN

def main() raises:
    var plot = (
        Plot()
        .mark_point()
        .encode(x=[1.0, 2.0, 3.0], y=[2.0, 5.0, 4.0])
        .labels(title="Styled Points")
        .theme(Theme(mark_color=SEAGREEN, point_radius=7.0))
    )
    save(plot, "styled.svg")

Categorical charts use the eight-color Okabe-Ito palette by default. Named alternatives live in dataviz.core.palettes:

from dataviz import Theme
from dataviz.core.palettes import tol_bright

var theme = Theme(categorical_palette=tol_bright())

tableau_10() and tol_muted() offer more categories, tol_light() supplies pale fills behind dark labels, and tab10_legacy() reproduces the previous default. A palette cycles when categories exceed its length, so use labels or another visual cue for larger sets.

Theme values cover canvas, axes, typography, marks, legends, annotations, and output. Presets provide coherent starting points: dark(), minimal(), high_contrast(), and print_safe(). Modify the returned Theme before attaching it when a preset needs an override.

See the theme recipes, Theme, and theme presets.