A silhouette plot visualizes the quality of clustering results by showing the silhouette coefficient for each sample, grouped by cluster assignment. Each horizontal bar represents a sample's silhouette score (-1 to 1), where positive values indicate good cluster membership and negative values suggest potential misclassification. This visualization helps evaluate cluster cohesion (how similar samples are to their own cluster) and separation (how distinct they are from neighboring clusters).
""" anyplot.ai
silhouette-basic: Silhouette Plot
Library: bokeh 3.9.0 | Python 3.13.13
Quality: 93/100 | Updated: 2026-05-10
"""
import os
import time
from pathlib import Path
import numpy as np
from bokeh.io import output_file, save
from bokeh.models import ColumnDataSource, Label, Span
from bokeh.plotting import figure
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
# Theme tokens
THEME = os.getenv("ANYPLOT_THEME", "light")
PAGE_BG = "#FAF8F1" if THEME == "light" else "#1A1A17"
INK = "#1A1A17" if THEME == "light" else "#F0EFE8"
INK_SOFT = "#4A4A44" if THEME == "light" else "#B8B7B0"
RULE = "rgba(26,26,23,0.10)" if THEME == "light" else "rgba(240,239,232,0.10)"
# Okabe-Ito colors for clusters
IMPRINT = ["#009E73", "#C475FD", "#4467A3"]
# Data - simulating silhouette analysis of customer segmentation (3 clusters)
# Realistic scenario: clustering customers by purchase behavior
np.random.seed(42)
n_clusters = 3
cluster_sizes = [50, 55, 45] # Different sized clusters
# Generate realistic silhouette values for each cluster
# Cluster 0: Well-separated cluster (high silhouette scores)
cluster0_vals = np.clip(np.random.beta(8, 2, cluster_sizes[0]) * 0.6 + 0.35, 0.1, 0.95)
# Cluster 1: Good cluster with some overlap (medium-high scores)
cluster1_vals = np.clip(np.random.beta(5, 2, cluster_sizes[1]) * 0.5 + 0.25, 0.0, 0.85)
# Cluster 2: Some ambiguous samples (includes negative values)
cluster2_vals = np.clip(np.random.beta(4, 3, cluster_sizes[2]) * 0.8 - 0.1, -0.15, 0.75)
# Combine all values
silhouette_vals = np.concatenate([cluster0_vals, cluster1_vals, cluster2_vals])
cluster_labels = np.concatenate(
[
np.zeros(cluster_sizes[0], dtype=int),
np.ones(cluster_sizes[1], dtype=int),
np.full(cluster_sizes[2], 2, dtype=int),
]
)
# Calculate average silhouette score
avg_silhouette = float(np.mean(silhouette_vals))
# Prepare data for plotting - sorted silhouette values within each cluster
y_lower = 15
bar_data = {"x": [], "y": [], "width": [], "height": [], "color": []}
cluster_info = [] # For labels
for i in range(n_clusters):
# Get silhouette values for this cluster
cluster_mask = cluster_labels == i
cluster_silhouette_vals = silhouette_vals[cluster_mask]
cluster_silhouette_vals.sort()
cluster_size = len(cluster_silhouette_vals)
y_upper = y_lower + cluster_size
# Store center position for cluster label
cluster_center = (y_lower + y_upper) / 2
cluster_avg = float(np.mean(cluster_silhouette_vals))
cluster_info.append((cluster_center, cluster_avg, cluster_size, i))
# Add bars for each sample in cluster
for j, val in enumerate(cluster_silhouette_vals):
bar_data["x"].append(val / 2) # Center of bar
bar_data["y"].append(y_lower + j + 0.5) # Y position
bar_data["width"].append(abs(val)) # Width = silhouette value
bar_data["height"].append(0.85) # Slightly less than 1 for gap
bar_data["color"].append(IMPRINT[i])
y_lower = y_upper + 15 # Gap between clusters
# Create figure
p = figure(
width=4800,
height=2700,
title="silhouette-basic · bokeh · anyplot.ai",
x_axis_label="Silhouette Coefficient",
y_axis_label="Cluster (samples sorted by silhouette score)",
x_range=(-0.3, 1.25),
y_range=(0, y_lower + 5),
tools="",
)
# Style the figure - sized for 4800x2700 canvas
p.title.text_font_size = "28pt"
p.xaxis.axis_label_text_font_size = "22pt"
p.yaxis.axis_label_text_font_size = "22pt"
p.xaxis.major_label_text_font_size = "18pt"
p.yaxis.major_label_text_font_size = "18pt"
p.xaxis.axis_label_standoff = 25
p.yaxis.axis_label_standoff = 25
# Theme-adaptive colors
p.background_fill_color = PAGE_BG
p.border_fill_color = PAGE_BG
p.outline_line_color = INK_SOFT
p.title.text_color = INK
p.xaxis.axis_label_text_color = INK
p.yaxis.axis_label_text_color = INK
p.xaxis.major_label_text_color = INK_SOFT
p.yaxis.major_label_text_color = INK_SOFT
p.xaxis.axis_line_color = INK_SOFT
p.yaxis.axis_line_color = INK_SOFT
p.xaxis.major_tick_line_color = INK_SOFT
p.yaxis.major_tick_line_color = INK_SOFT
# Create data source for bars
source = ColumnDataSource(data=bar_data)
# Draw horizontal bars
p.rect(x="x", y="y", width="width", height="height", color="color", source=source, line_color=None, alpha=0.85)
# Add vertical line for average silhouette score
avg_line = Span(location=avg_silhouette, dimension="height", line_color=INK_SOFT, line_width=3, line_dash="dashed")
p.add_layout(avg_line)
# Add average silhouette score label at top
avg_label = Label(
x=avg_silhouette + 0.03,
y=y_lower - 5,
text=f"Average: {avg_silhouette:.3f}",
text_font_size="20pt",
text_color=INK_SOFT,
text_font_style="bold",
)
p.add_layout(avg_label)
# Add cluster labels with their average silhouette scores
for center_y, cluster_avg, size, cluster_idx in cluster_info:
# Position label to the left side, outside the bars
cluster_label = Label(
x=-0.22,
y=center_y,
text=f"Cluster {cluster_idx}",
text_font_size="18pt",
text_color=IMPRINT[cluster_idx],
text_font_style="bold",
text_align="left",
text_baseline="middle",
)
p.add_layout(cluster_label)
# Add cluster stats on the right side
stats_label = Label(
x=1.01,
y=center_y,
text=f"n={size}, avg={cluster_avg:.2f}",
text_font_size="16pt",
text_color=IMPRINT[cluster_idx],
text_font_style="normal",
text_align="left",
text_baseline="middle",
)
p.add_layout(stats_label)
# Style grid
p.xgrid.grid_line_color = RULE
p.xgrid.grid_line_alpha = 0.10
p.xgrid.grid_line_dash = [6, 4]
p.ygrid.grid_line_alpha = 0.0 # No horizontal grid
# Remove y-axis ticks (sample indices are not meaningful)
p.yaxis.major_tick_line_color = None
p.yaxis.minor_tick_line_color = None
p.yaxis.major_label_text_font_size = "0pt"
# Add vertical line at x=0 for reference
zero_line = Span(location=0, dimension="height", line_color=INK_SOFT, line_width=2, line_alpha=0.5)
p.add_layout(zero_line)
# Save as HTML first
output_file(f"plot-{THEME}.html")
save(p)
# Screenshot with headless Chrome
W, H = 4800, 2700
opts = Options()
for arg in (
"--headless=new",
"--no-sandbox",
"--disable-dev-shm-usage",
"--disable-gpu",
f"--window-size={W},{H}",
"--hide-scrollbars",
):
opts.add_argument(arg)
driver = webdriver.Chrome(options=opts)
driver.set_window_size(W, H)
driver.get(f"file://{Path(f'plot-{THEME}.html').resolve()}")
time.sleep(3)
driver.save_screenshot(f"plot-{THEME}.png")
driver.quit()
Part of Silhouette Plot on anyplot.ai.