---
id: graspologic
version: "3.4.4"
license: MIT
license_treatment: permissive
maintenance: active
---

# graspologic — A set of Python modules for graph statistics

License: permissive · Maintenance: active · Downloads: 148.2K/mo

## What it is and what it does

Graspologic is a Python library for statistical analysis of graphs and networks. It provides algorithms and utilities designed to work with the spatial structure of networks, applying specialized statistical techniques rather than treating graph data as unstructured. The package includes methods for graph embedding, clustering, hypothesis testing, and other analyses that respect the relational structure inherent in network data.

The library depends on a mature scientific Python stack (numpy, scipy, scikit-learn, networkx, matplotlib, seaborn) and is actively maintained with support for Python 3.9 through 3.12. It is intended for researchers and practitioners working with network data in domains like social networks, biological networks, or any domain where relationships between entities matter. The package is production-stable and has been in active development since 2020.

Use it for:

- Embed graph structures into low-dimensional vector spaces for downstream machine learning tasks.
- Test statistical hypotheses about network structure, connectivity, or community organization.
- Cluster nodes or detect communities within networks using graph-aware algorithms.
- Analyze and compare multiple networks to identify structural differences or similarities.
- Preprocess and visualize network data for exploratory analysis or publication.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Graspologic provides graph statistical algorithms and utilities for processing, analyzing, and modeling networks using specialized statistical techniques that account for graph structure.

Yes. Graspologic is a mature, actively maintained library with no known vulnerabilities, permissive licensing, and low installation friction. It is well-suited for anyone working with network or graph data who needs statistical algorithms beyond basic graph operations. The dependency footprint is substantial but standard in scientific Python; if you already use scikit-learn and scipy, adding graspologic is straightforward.

## Install

pip install graspologic
uv add graspologic
poetry add graspologic

## Installing graspologic

Before you install: Low friction installation with a pure-Python wheel. Active maintenance with a recent release (340 days ago) and ongoing repository activity. Depends on 17 runtime packages including heavy scientific libraries (scikit-learn, scipy, networkx, gensim), which are standard in the data science ecosystem.

License in practice: MIT license is permissive—you can use, modify, and distribute graspologic freely in commercial and private projects with minimal restrictions, only requiring license attribution.

Quickstart:

pip install graspologic

import graspologic
from graspologic.embed import AdjacencySpectralEmbed

# Embed a graph adjacency matrix
embedder = AdjacencySpectralEmbed()
embedding = embedder.fit_transform(adjacency_matrix)

Requires Python 3.9–3.12 (x86_64 architecture on Linux, macOS, or Windows 10); tested on these platforms only.

Verify before relying:

- Whether the package's graph algorithms scale well to very large networks or if there are documented size/performance limits.
- Whether the embedding and clustering methods are suitable for directed, weighted, or temporal graphs specifically.
- Performance characteristics when working with sparse versus dense adjacency matrices.

## Package facts

- License: MIT (permissive)
- Python support: capped_below_current
- Install friction: low
- Maintenance: active
- Downloads: 148.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

graph statistics python, network analysis algorithms, graph statistical methods, network modeling tools, graph embedding and clustering, statistical graph processing, network structure analysis, graph-analysis, network-science, statistical-learning

[View on SkillFed](https://skillfed.io/packages/graspologic) · [View on PyPI](https://pypi.org/project/graspologic/)