An UpSet plot visualizes intersections of multiple sets using a matrix-based layout that scales far better than Venn diagrams beyond 3 sets. A horizontal bar chart shows individual set sizes, a dot-matrix indicates which sets participate in each intersection, and a vertical bar chart above shows the intersection cardinality. This is the modern standard for set intersection analysis, making complex overlaps between many sets immediately readable.
""" anyplot.ai
upset-basic: UpSet Plot for Multi-Set Intersection Analysis
Library: seaborn 0.13.2 | Python 3.13.13
Quality: 82/100 | Created: 2026-05-13
"""
import os
from collections import Counter
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
# 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"
BRAND = "#009E73"
sns.set_theme(
style="ticks",
rc={
"figure.facecolor": PAGE_BG,
"axes.facecolor": PAGE_BG,
"axes.edgecolor": INK_SOFT,
"axes.labelcolor": INK,
"text.color": INK,
"xtick.color": INK_SOFT,
"ytick.color": INK_SOFT,
"grid.color": INK,
"grid.alpha": 0.10,
"legend.facecolor": ELEVATED_BG,
"legend.edgecolor": INK_SOFT,
},
)
# Data: customer segments across marketing channels
np.random.seed(42)
n_customers = 1200
channel_names = ["Email", "Mobile", "Social", "Premium", "Loyalty"]
channel_probs = [0.62, 0.51, 0.39, 0.27, 0.21]
membership = np.column_stack([np.random.random(n_customers) < p for p in channel_probs])
membership = membership[membership.any(axis=1)]
n_sets = len(channel_names)
intersection_counts = Counter(tuple(row) for row in membership)
sorted_ints = sorted(intersection_counts.items(), key=lambda x: -x[1])
top_n = min(14, len(sorted_ints))
top_keys = [k for k, _ in sorted_ints[:top_n]]
top_counts = [v for _, v in sorted_ints[:top_n]]
n_ints = len(top_keys)
set_sizes = membership.sum(axis=0)
# Figure layout
fig = plt.figure(figsize=(16, 9), facecolor=PAGE_BG)
gs = fig.add_gridspec(
2,
2,
width_ratios=[1.4, 5],
height_ratios=[3, 2],
hspace=0.0,
wspace=0.0,
left=0.07,
right=0.97,
top=0.88,
bottom=0.05,
)
ax_corner = fig.add_subplot(gs[0, 0])
ax_intbar = fig.add_subplot(gs[0, 1])
ax_setbar = fig.add_subplot(gs[1, 0])
ax_matrix = fig.add_subplot(gs[1, 1])
x_cols = np.arange(n_ints)
y_rows = np.arange(n_sets)
# Intersection bar chart (top right)
ax_intbar.set_facecolor(PAGE_BG)
ax_intbar.set_axisbelow(True)
ax_intbar.yaxis.grid(True, alpha=0.10, linewidth=0.8, color=INK)
ax_intbar.bar(x_cols, top_counts, color=BRAND, width=0.55, zorder=2)
for i, cnt in enumerate(top_counts):
ax_intbar.text(i, cnt + max(top_counts) * 0.015, str(cnt), ha="center", va="bottom", fontsize=10, color=INK_SOFT)
ax_intbar.set_xlim(-0.5, n_ints - 0.5)
ax_intbar.set_ylabel("Intersection Size", fontsize=17, color=INK, labelpad=8)
ax_intbar.tick_params(axis="y", labelsize=13, colors=INK_SOFT)
ax_intbar.tick_params(axis="x", bottom=False, labelbottom=False)
ax_intbar.spines["top"].set_visible(False)
ax_intbar.spines["right"].set_visible(False)
ax_intbar.spines["bottom"].set_visible(False)
ax_intbar.spines["left"].set_color(INK_SOFT)
# Dot matrix (bottom right)
ax_matrix.set_facecolor(PAGE_BG)
for row_idx in range(n_sets):
ax_matrix.axhline(row_idx, color=INK, alpha=0.06, linewidth=0.8, zorder=1)
for col_idx, key in enumerate(top_keys):
active_rows = [r for r, active in enumerate(key) if active]
for row_idx, is_active in enumerate(key):
if is_active:
ax_matrix.scatter(col_idx, row_idx, s=200, color=BRAND, zorder=4, linewidths=0)
else:
ax_matrix.scatter(col_idx, row_idx, s=55, color=INK_MUTED, alpha=0.28, zorder=3, linewidths=0)
if len(active_rows) > 1:
ax_matrix.plot(
[col_idx, col_idx],
[min(active_rows), max(active_rows)],
color=BRAND,
linewidth=3.5,
zorder=3,
solid_capstyle="round",
)
ax_matrix.set_xlim(-0.5, n_ints - 0.5)
ax_matrix.set_ylim(-0.5, n_sets - 0.5)
ax_matrix.set_yticks(y_rows)
ax_matrix.set_yticklabels(channel_names, fontsize=14, color=INK)
ax_matrix.tick_params(axis="x", bottom=False, labelbottom=False)
ax_matrix.tick_params(axis="y", left=False)
for spine in ax_matrix.spines.values():
spine.set_visible(False)
# Set size bars (bottom left, bars extend left toward margin)
ax_setbar.set_facecolor(PAGE_BG)
ax_setbar.set_axisbelow(True)
ax_setbar.xaxis.grid(True, alpha=0.10, linewidth=0.8, color=INK)
for row_idx in range(n_sets):
ax_setbar.axhline(row_idx, color=INK, alpha=0.06, linewidth=0.8, zorder=1)
ax_setbar.barh(y_rows, set_sizes, color=BRAND, height=0.45, zorder=2)
ax_setbar.set_ylim(-0.5, n_sets - 0.5)
ax_setbar.invert_xaxis()
ax_setbar.set_xlabel("Set Size", fontsize=15, color=INK, labelpad=6)
ax_setbar.tick_params(axis="x", labelsize=12, colors=INK_SOFT)
ax_setbar.tick_params(axis="y", left=False, labelleft=False)
ax_setbar.spines["top"].set_visible(False)
ax_setbar.spines["left"].set_visible(False)
ax_setbar.spines["right"].set_visible(False)
ax_setbar.spines["bottom"].set_color(INK_SOFT)
# Empty corner
ax_corner.set_facecolor(PAGE_BG)
ax_corner.tick_params(left=False, bottom=False, labelleft=False, labelbottom=False)
for spine in ax_corner.spines.values():
spine.set_visible(False)
# Title
fig.text(
0.5,
0.93,
"Customer Segments · upset-basic · seaborn · anyplot.ai",
ha="center",
va="center",
fontsize=21,
fontweight="medium",
color=INK,
)
# Save
plt.savefig(f"plot-{THEME}.png", dpi=300, bbox_inches="tight", facecolor=PAGE_BG)
Part of UpSet Plot for Multi-Set Intersection Analysis on anyplot.ai.