A mosaic plot visualizes contingency tables by dividing a rectangular area into smaller rectangles whose areas are proportional to cell frequencies. This statistical visualization technique effectively shows relationships and associations between two or more categorical variables, making it easy to identify patterns, dependencies, and deviations from expected frequencies in cross-tabulated data.
""" anyplot.ai
mosaic-categorical: Mosaic Plot for Categorical Association Analysis
Library: pygal 3.1.0 | Python 3.13.13
Quality: 89/100 | Updated: 2026-05-19
"""
import os
import sys
import numpy as np
# Theme tokens
THEME = os.getenv("ANYPLOT_THEME", "light")
PAGE_BG = "#FAF8F1" if THEME == "light" else "#1A1A17"
ELEVATED_BG = "#FFFDF6" if THEME == "light" else "#242420"
INK = "#1A1A17" if THEME == "light" else "#F0EFE8"
INK_SOFT = "#4A4A44" if THEME == "light" else "#B8B7B0"
INK_MUTED = "#6B6A63" if THEME == "light" else "#A8A79F"
RULE = "rgba(26,26,23,0.15)" if THEME == "light" else "rgba(240,239,232,0.15)"
# Okabe-Ito palette — first series is always brand green
IMPRINT = ("#009E73", "#C475FD", "#4467A3", "#BD8233")
# Temporarily remove current directory from path to avoid pygal.py name collision
_cwd = sys.path[0] if sys.path[0] else "."
if _cwd in sys.path:
sys.path.remove(_cwd)
from pygal.graph.graph import Graph
from pygal.style import Style
sys.path.insert(0, _cwd)
class MosaicPlot(Graph):
def __init__(self, *args, **kwargs):
self.contingency_data = kwargs.pop("contingency_data", {})
self.row_labels = kwargs.pop("row_labels", [])
self.col_labels = kwargs.pop("col_labels", [])
self.cell_colors = kwargs.pop("cell_colors", list(IMPRINT))
self.gap_ratio = kwargs.pop("gap_ratio", 0.02)
self.ink = kwargs.pop("ink", INK)
self.ink_soft = kwargs.pop("ink_soft", INK_SOFT)
self.ink_muted = kwargs.pop("ink_muted", INK_MUTED)
self.page_bg = kwargs.pop("page_bg", PAGE_BG)
self.rule = kwargs.pop("rule", RULE)
super().__init__(*args, **kwargs)
def _plot(self):
if not self.contingency_data or not self.row_labels or not self.col_labels:
return
n_rows = len(self.row_labels)
n_cols = len(self.col_labels)
col_totals = [
sum(self.contingency_data.get((row, col), 0) for row in self.row_labels) for col in self.col_labels
]
grand_total = sum(col_totals)
if grand_total == 0:
return
col_proportions = [t / grand_total for t in col_totals]
plot_width = self.view.width
plot_height = self.view.height
margin_left = 440
margin_right = 80
margin_top = 100
margin_bottom = 380
available_width = plot_width - margin_left - margin_right
available_height = plot_height - margin_top - margin_bottom
col_gap = available_width * self.gap_ratio
row_gap = available_height * self.gap_ratio
drawing_width = available_width - col_gap * (n_cols - 1)
drawing_height = available_height - row_gap * (n_rows - 1)
x_offset = self.view.x(0) + margin_left
y_offset = self.view.y(n_rows) + margin_top
plot_node = self.nodes["plot"]
mosaic_group = self.svg.node(plot_node, class_="mosaic-plot")
current_x = x_offset
for j, col in enumerate(self.col_labels):
col_width = drawing_width * col_proportions[j]
col_total = col_totals[j]
if col_total == 0:
current_x += col_width + col_gap
continue
row_values = [self.contingency_data.get((row, col), 0) for row in self.row_labels]
row_proportions = [v / col_total for v in row_values]
current_y = y_offset
for i, row in enumerate(self.row_labels):
cell_height = drawing_height * row_proportions[i]
if cell_height < 1:
current_y += cell_height + row_gap
continue
color = self.cell_colors[i % len(self.cell_colors)]
freq = self.contingency_data.get((row, col), 0)
pct = row_proportions[i] * 100
rect = self.svg.node(
mosaic_group,
"rect",
x=current_x,
y=current_y,
width=max(col_width, 1),
height=max(cell_height, 1),
rx=6,
ry=6,
)
rect.set("fill", color)
rect.set("stroke", self.page_bg)
rect.set("stroke-width", "5")
rect.set("fill-opacity", "0.90")
# SVG tooltip for interactive HTML — pygal-native feature
tooltip = self.svg.node(rect, "title")
tooltip.text = f"{row} · {col}: {freq:,} ({pct:.1f}% of {col} users)"
if cell_height > 80 and col_width > 110:
label_size = min(44, int(min(col_width, cell_height) * 0.26))
text_x = current_x + col_width / 2
if cell_height > 130 and col_width > 160:
# Dual label: count (bold, upper) + percentage (smaller, lower)
count_node = self.svg.node(
mosaic_group, "text", x=text_x, y=current_y + cell_height / 2 - label_size * 0.28
)
count_node.set("text-anchor", "middle")
count_node.set("dominant-baseline", "middle")
count_node.set("fill", "#ffffff")
count_node.set(
"style",
f"font-size:{label_size}px;font-weight:700;font-family:sans-serif;letter-spacing:-0.5px",
)
count_node.text = f"{freq:,}"
pct_size = max(int(label_size * 0.62), 22)
pct_node = self.svg.node(
mosaic_group, "text", x=text_x, y=current_y + cell_height / 2 + label_size * 0.52
)
pct_node.set("text-anchor", "middle")
pct_node.set("dominant-baseline", "middle")
pct_node.set("fill", "rgba(255,255,255,0.80)")
pct_node.set("style", f"font-size:{pct_size}px;font-family:sans-serif;font-weight:500")
pct_node.text = f"{pct:.0f}%"
else:
# Single count label for smaller cells
count_node = self.svg.node(mosaic_group, "text", x=text_x, y=current_y + cell_height / 2)
count_node.set("text-anchor", "middle")
count_node.set("dominant-baseline", "middle")
count_node.set("fill", "#ffffff")
count_node.set("style", f"font-size:{label_size}px;font-weight:700;font-family:sans-serif")
count_node.text = f"{freq:,}"
current_y += cell_height + row_gap
# Column label — bold, prominent
col_label_size = 46
label_x = current_x + col_width / 2
label_y = y_offset + available_height + 76
col_label_node = self.svg.node(mosaic_group, "text", x=label_x, y=label_y)
col_label_node.set("text-anchor", "middle")
col_label_node.set("fill", self.ink)
col_label_node.set(
"style", f"font-size:{col_label_size}px;font-weight:700;font-family:sans-serif;letter-spacing:0.5px"
)
col_label_node.text = col
# Hairline separator between label and proportion sublabel
sep_y = label_y + 16
half_w = min(col_width * 0.35, 100)
sep_line = self.svg.node(mosaic_group, "line", x1=label_x - half_w, y1=sep_y, x2=label_x + half_w, y2=sep_y)
sep_line.set("stroke", self.ink_muted)
sep_line.set("stroke-width", "1.5")
sep_line.set("stroke-opacity", "0.45")
# Proportion sub-label — italic, muted
prop_label_y = label_y + 58
prop_node = self.svg.node(mosaic_group, "text", x=label_x, y=prop_label_y)
prop_node.set("text-anchor", "middle")
prop_node.set("fill", self.ink_muted)
prop_node.set("style", "font-size:34px;font-style:italic;font-family:sans-serif")
prop_node.text = f"{col_proportions[j] * 100:.1f}% of sessions"
current_x += col_width + col_gap
# Row legend (left side) — larger swatches, refined alignment
first_col = self.col_labels[0]
first_col_total = col_totals[0] if col_totals[0] > 0 else 1
for i, row in enumerate(self.row_labels):
color = self.cell_colors[i % len(self.cell_colors)]
cumulative = sum(
self.contingency_data.get((r, first_col), 0) / first_col_total for r in self.row_labels[:i]
)
row_prop = self.contingency_data.get((row, first_col), 0) / first_col_total
center_y = y_offset + drawing_height * (cumulative + row_prop / 2) + i * row_gap
swatch_size = 40
swatch_x = x_offset - 260
swatch_y = center_y - swatch_size / 2
swatch = self.svg.node(
mosaic_group, "rect", x=swatch_x, y=swatch_y, width=swatch_size, height=swatch_size, rx=6, ry=6
)
swatch.set("fill", color)
swatch.set("stroke", self.ink_soft)
swatch.set("stroke-width", "1.5")
swatch.set("fill-opacity", "0.90")
row_label_size = 38
text_x = swatch_x + swatch_size + 16
text_y = center_y + row_label_size * 0.36
row_label_node = self.svg.node(mosaic_group, "text", x=text_x, y=text_y)
row_label_node.set("text-anchor", "start")
row_label_node.set("fill", self.ink)
row_label_node.set("style", f"font-size:{row_label_size}px;font-family:sans-serif;font-weight:500")
row_label_node.text = row
# X-axis title
x_title_x = x_offset + available_width / 2
x_title_y = y_offset + available_height + 216
x_title_node = self.svg.node(mosaic_group, "text", x=x_title_x, y=x_title_y)
x_title_node.set("text-anchor", "middle")
x_title_node.set("fill", self.ink_soft)
x_title_node.set("style", "font-size:50px;font-weight:700;font-family:sans-serif;letter-spacing:1px")
x_title_node.text = "DEVICE TYPE"
# Y-axis title (rotated)
y_title_x = x_offset - 400
y_title_y = y_offset + available_height / 2
y_title_node = self.svg.node(mosaic_group, "text", x=y_title_x, y=y_title_y)
y_title_node.set("text-anchor", "middle")
y_title_node.set("fill", self.ink_soft)
y_title_node.set("style", "font-size:50px;font-weight:700;font-family:sans-serif;letter-spacing:1px")
y_title_node.set("transform", f"rotate(-90, {y_title_x}, {y_title_y})")
y_title_node.text = "APP CATEGORY"
# Key insight annotation — guides viewer to the most striking pattern
insight_y = y_offset + available_height + 318
insight_node = self.svg.node(mosaic_group, "text", x=x_title_x, y=insight_y)
insight_node.set("text-anchor", "middle")
insight_node.set("fill", self.ink_muted)
insight_node.set("style", "font-size:34px;font-style:italic;font-family:sans-serif")
insight_node.text = (
"Key insight: Desktop skews heavily towards Productivity (57%) — nearly 4× the Mobile rate (15%)"
)
def _compute(self):
n_rows = len(self.row_labels) if self.row_labels else 1
n_cols = len(self.col_labels) if self.col_labels else 1
self._box.xmin = 0
self._box.xmax = n_cols
self._box.ymin = 0
self._box.ymax = n_rows
# App engagement data: device type vs. app category
np.random.seed(42)
data = {
("Social Media", "Mobile"): 450,
("Gaming", "Mobile"): 320,
("Productivity", "Mobile"): 180,
("Entertainment", "Mobile"): 290,
("Social Media", "Tablet"): 120,
("Gaming", "Tablet"): 200,
("Productivity", "Tablet"): 230,
("Entertainment", "Tablet"): 290,
("Social Media", "Desktop"): 80,
("Gaming", "Desktop"): 150,
("Productivity", "Desktop"): 480,
("Entertainment", "Desktop"): 130,
}
row_labels = ["Social Media", "Gaming", "Productivity", "Entertainment"]
col_labels = ["Mobile", "Tablet", "Desktop"]
custom_style = Style(
background=PAGE_BG,
plot_background=PAGE_BG,
foreground=INK,
foreground_strong=INK,
foreground_subtle=INK_MUTED,
colors=IMPRINT,
title_font_size=68,
legend_font_size=40,
label_font_size=40,
value_font_size=32,
font_family="sans-serif",
)
chart = MosaicPlot(
width=4800,
height=2700,
style=custom_style,
title="App Usage by Device · mosaic-categorical · python · pygal · anyplot.ai",
contingency_data=data,
row_labels=row_labels,
col_labels=col_labels,
cell_colors=list(IMPRINT),
gap_ratio=0.015,
show_legend=False,
margin=100,
margin_top=220,
margin_bottom=150,
show_x_labels=False,
show_y_labels=False,
ink=INK,
ink_soft=INK_SOFT,
ink_muted=INK_MUTED,
page_bg=PAGE_BG,
rule=RULE,
)
# Dummy series required to trigger pygal's rendering pipeline
chart.add("", [0])
# Save
chart.render_to_png(f"plot-{THEME}.png")
with open(f"plot-{THEME}.html", "wb") as f:
f.write(chart.render())
Part of Mosaic Plot for Categorical Association Analysis on anyplot.ai.