A dendrogram visualizes hierarchical clustering by showing how data points or clusters merge at different distance levels. The tree-like structure reveals relationships and similarity between items, with branch heights indicating the distance at which clusters merge. This visualization is essential for understanding the hierarchical structure in data and identifying natural groupings.
""" anyplot.ai
dendrogram-basic: Basic Dendrogram
Library: letsplot 4.10.1 | Python 3.13.13
Quality: 87/100 | Updated: 2026-06-18
"""
import os
import numpy as np
import pandas as pd
from lets_plot import (
LetsPlot,
aes,
element_blank,
element_line,
element_rect,
element_text,
geom_hline,
geom_point,
geom_segment,
geom_text,
ggplot,
ggsize,
labs,
layer_tooltips,
scale_color_manual,
scale_x_continuous,
scale_y_continuous,
theme,
theme_void,
)
from lets_plot.export import ggsave
from scipy.cluster.hierarchy import linkage
from sklearn.datasets import load_iris
LetsPlot.setup_html()
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"
# Imprint palette — positions 1-4 in canonical order
CLUSTER_COLORS = ["#009E73", "#C475FD", "#4467A3", "#BD8233"]
CLUSTER_BREAKS = ["Setosa", "Versicolor", "Virginica", "Cross-cluster"]
# Data — Iris flower measurements (15 samples, 3 species)
iris = load_iris()
np.random.seed(42)
indices = np.sort(np.concatenate([np.random.choice(np.where(iris.target == k)[0], 5, replace=False) for k in range(3)]))
features = iris.data[indices]
species_names = ["Setosa", "Versicolor", "Virginica"]
labels = [f"{species_names[iris.target[i]][:3]}-{j + 1}" for j, i in enumerate(indices)]
# Hierarchical clustering (Ward's method)
linkage_matrix = linkage(features, method="ward")
# Build dendrogram segment coordinates from linkage matrix
n = len(labels)
leaf_positions = {i: float(i) for i in range(n)}
node_heights = dict.fromkeys(range(n), 0.0)
segments = []
# Color threshold at 70% of max distance — splits into 3 major species clusters
max_dist = linkage_matrix[:, 2].max()
color_threshold = 0.7 * max_dist
prefix_to_species = {"Set": "Setosa", "Ver": "Versicolor", "Vir": "Virginica"}
node_cluster = {i: prefix_to_species[labels[i].split("-")[0]] for i in range(n)}
for i, (left, right, dist, _) in enumerate(linkage_matrix):
left, right = int(left), int(right)
new_node = n + i
left_pos = leaf_positions[left]
right_pos = leaf_positions[right]
leaf_positions[new_node] = (left_pos + right_pos) / 2
node_heights[new_node] = dist
left_cl, right_cl = node_cluster[left], node_cluster[right]
node_cluster[new_node] = left_cl if left_cl == right_cl else "Cross-cluster"
cluster_label = node_cluster[new_node] if dist < color_threshold else "Cross-cluster"
lh, rh = node_heights[left], node_heights[right]
for seg in [(left_pos, lh, left_pos, dist), (right_pos, rh, right_pos, dist), (left_pos, dist, right_pos, dist)]:
segments.append(
{
"x": seg[0],
"y": seg[1],
"xend": seg[2],
"yend": seg[3],
"cluster": cluster_label,
"merge_dist": round(dist, 2),
}
)
segment_df = pd.DataFrame(segments)
leaf_data = [
{"x": leaf_positions[i], "y": 0, "label": labels[i], "cluster": prefix_to_species[labels[i].split("-")[0]]}
for i in range(n)
]
label_df = pd.DataFrame(leaf_data)
plot = (
ggplot()
+ geom_segment(
aes(x="x", y="y", xend="xend", yend="yend", color="cluster"),
data=segment_df,
size=1.5,
tooltips=layer_tooltips().title("@cluster").line("Merge distance|@merge_dist").min_width(180),
)
+ geom_point(
aes(x="x", y="y", color="cluster"),
data=label_df,
size=2.5,
shape=16,
show_legend=False,
tooltips=layer_tooltips().title("@cluster").line("Sample|@label"),
)
+ geom_text(
aes(x="x", y="y", label="label", color="cluster"),
data=label_df.assign(y=-max_dist * 0.05),
angle=45,
hjust=1,
vjust=1,
size=4,
family="monospace",
show_legend=False,
)
+ geom_hline(yintercept=color_threshold, linetype="dashed", color=INK_MUTED, size=0.8)
+ geom_text(
aes(x="x", y="y", label="label"),
data=pd.DataFrame(
[{"x": n - 1.8, "y": color_threshold + max_dist * 0.03, "label": f"threshold = {color_threshold:.1f}"}]
),
size=3.5,
color=INK_MUTED,
hjust=1,
family="monospace",
)
+ scale_color_manual(values=CLUSTER_COLORS, breaks=CLUSTER_BREAKS, name="Cluster")
+ scale_x_continuous(expand=[0.06, 0.02])
+ scale_y_continuous(
name="Ward Linkage Distance",
limits=[-max_dist * 0.14, max_dist * 1.07],
expand=[0, 0],
breaks=[0, 2, 4, 6, 8, 10, 12],
)
+ labs(x="", title="dendrogram-basic · python · letsplot · anyplot.ai")
+ theme_void()
+ theme(
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
plot_title=element_text(size=16, face="bold", color=INK),
axis_title_y=element_text(size=12, color=INK_SOFT, margin=[0, 8, 0, 0]),
axis_text_y=element_text(size=10, color=INK_SOFT),
axis_text_x=element_blank(),
axis_ticks_x=element_blank(),
axis_ticks_y=element_line(size=0.4, color=INK_SOFT),
axis_line_y=element_line(size=0.6, color=INK_SOFT),
panel_grid_major_y=element_line(size=0.3, color=INK_MUTED),
legend_background=element_rect(fill=ELEVATED_BG, color=INK_SOFT),
legend_text=element_text(size=10, color=INK_SOFT),
legend_title=element_text(size=10, color=INK),
plot_margin=[25, 15, 20, 10],
)
+ ggsize(800, 450)
)
ggsave(plot, f"plot-{THEME}.png", path=".", scale=4)
ggsave(plot, f"plot-{THEME}.html", path=".")
Part of Basic Dendrogram on anyplot.ai.