linkage2df 1.0.0

Convert SciPy linkage matrices to plot-ready DataFrames — used for ggplot/Plotnine dendrograms.

linkage2df

Easily plot SciPy dendrograms using the Grammar of Graphics in Python

linkage2df is a tool that converts SciPy linkage matrices (typically used for hierarchical clustering) into tidy DataFrames. By transforming the dendrogram tree into a set of segments with (x, y) coordinates, you can leverage the full power of Plotnine (or any ggplot2 clone) to create highly customizable dendrograms.

Features

• Tidy Data Output: Converts hierarchical trees into a simple DataFrame of segments with x_start, y_start, x_end, and y_end columns.
• Cluster Propagation: Automatically assigns clusters to higher-level branches based on their children, allowing for intuitive coloring of the tree.
• Plotnine Ready: Designed specifically to work with geom_segment() in the Grammar of Graphics.

---

Installation

You can install the package directly from your local directory:

pip install .

Quick start

import numpy as np
import pandas as pd
from scipy.cluster.hierarchy import linkage, fcluster
from plotnine import ggplot, aes, geom_segment, theme_minimal
from linkage2df import linkage_to_segments

# 1. Generate some dummy data
np.random.seed(42)
X = np.random.rand(15, 4)

# 2. Compute linkage matrix & clusters
linkage_matrix = linkage(X, method="ward")
clusters = fcluster(linkage_matrix, t=0.7, criterion='distance')

# 3. Convert linkage matrix to a tidy segments DataFrame
df = linkage_to_segments(linkage_matrix, clusters=clusters)

# 4. Plot using Plotnine
p = (
    ggplot(df) +
    geom_segment(aes(x='x_start', y='y_start', xend='x_end', yend='y_end'), size=0.75) +
    theme_minimal()
)

print(p)