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: altair 6.1.0 | Python 3.13.13
Quality: 62/100 | Updated: 2026-05-10
"""
import altair as alt
import numpy as np
import pandas as pd
from sklearn.cluster import KMeans
from sklearn.datasets import load_iris
from sklearn.metrics import silhouette_samples, silhouette_score
# Data - Use iris dataset for realistic clustering example
np.random.seed(42)
iris = load_iris()
X = iris.data
# Perform K-means clustering with 3 clusters
n_clusters = 3
kmeans = KMeans(n_clusters=n_clusters, random_state=42, n_init=10)
cluster_labels = kmeans.fit_predict(X)
# Calculate silhouette scores
silhouette_avg = silhouette_score(X, cluster_labels)
sample_silhouette_values = silhouette_samples(X, cluster_labels)
# Build dataframe with samples sorted by cluster and silhouette score
# Horizontal bars: y-axis is sample position, x-axis is silhouette score
records = []
y_lower = 0
for cluster_id in range(n_clusters):
cluster_mask = cluster_labels == cluster_id
cluster_silhouette_values = sample_silhouette_values[cluster_mask]
cluster_silhouette_values.sort()
cluster_size = len(cluster_silhouette_values)
cluster_avg = np.mean(cluster_silhouette_values)
for i, sil_value in enumerate(cluster_silhouette_values):
records.append(
{
"y": y_lower + i,
"y2": y_lower + i + 0.8, # Bar thickness
"silhouette": sil_value,
"cluster": f"Cluster {cluster_id} (avg: {cluster_avg:.2f})",
"cluster_id": cluster_id,
}
)
y_lower += cluster_size + 5 # Gap between clusters
df = pd.DataFrame(records)
# Color palette - Python Blue as primary, then distinct colors
colors = ["#306998", "#FFD43B", "#E34C26"]
# Create horizontal silhouette bars using mark_rect for proper horizontal bars
bars = (
alt.Chart(df)
.mark_rect()
.encode(
x=alt.X("silhouette:Q", title="Silhouette Coefficient", scale=alt.Scale(domain=[-0.15, 1.0])),
x2=alt.value(0), # Bars start from 0 and extend to silhouette value
y=alt.Y("y:Q", axis=None),
y2="y2:Q",
color=alt.Color(
"cluster:N",
title="Cluster",
scale=alt.Scale(domain=df["cluster"].unique().tolist(), range=colors),
legend=alt.Legend(titleFontSize=18, labelFontSize=16, symbolSize=200),
),
tooltip=[
alt.Tooltip("cluster:N", title="Cluster"),
alt.Tooltip("silhouette:Q", title="Silhouette Score", format=".3f"),
],
)
)
# Vertical line for average silhouette score
avg_line_data = pd.DataFrame({"avg_silhouette": [silhouette_avg]})
avg_line = (
alt.Chart(avg_line_data)
.mark_rule(color="#E63946", strokeWidth=3, strokeDash=[8, 4])
.encode(
x=alt.X("avg_silhouette:Q"), tooltip=[alt.Tooltip("avg_silhouette:Q", title="Average Silhouette", format=".3f")]
)
)
# Annotation for average line
avg_text = (
alt.Chart(pd.DataFrame({"x": [silhouette_avg + 0.02], "y": [5], "text": [f"Avg: {silhouette_avg:.3f}"]}))
.mark_text(fontSize=18, fontWeight="bold", color="#E63946", align="left")
.encode(x="x:Q", y="y:Q", text="text:N")
)
# Combine layers
chart = (
alt.layer(bars, avg_line, avg_text)
.properties(width=1600, height=900, title=alt.Title("silhouette-basic · altair · pyplots.ai", fontSize=28))
.configure_axis(labelFontSize=18, titleFontSize=22)
.configure_legend(titleFontSize=18, labelFontSize=16)
)
# Save as PNG (4800x2700 at scale_factor=3)
chart.save("plot.png", scale_factor=3.0)
# Save as HTML for interactivity
chart.save("plot.html")
Part of Silhouette Plot on anyplot.ai.