mvcluster 1.17

A Python package for multiview unsupervised clustering

mvcluster: Multiview Clustering Python Package

• mvcluster: Multiview Clustering Python Package
  • Project Overview
  • Key Features
  • Installation
  • Requirements:
  • Quick Start
  • Documentation
  • Research Context
  • Contributing
  • Citation
  • Contact

---

Project Overview

mvcluster is an open-source Python package developed during my research internship at Centre Borelli (Université Paris Cité) from April to July 2025. The package implements advanced algorithms for multiview clustering, with a focus on graph-structured data.

Key Features

• Implementation of LMGEC (Linear Multiview Graph Embedding and Clustering) algorithm
• Standardized scikit-learn compatible API
• Comprehensive documentation and examples
• Extensive test coverage (>90%)
• Support for multiple multiview datasets
• Visualization tools for clustering results

Installation

pip install mvcluster

Requirements:

• Python ≥ 3.8

• numpy ≥ 1.21

• scipy ≥ 1.7

• scikit-learn ≥ 1.0

• torch ≥ 1.9.0 (optional for GPU acceleration)

• networkx ≥ 2.6

Quick Start

from mvcluster.cluster import LMGEC
from mvcluster.utils.datagen import datagen
# Load sample dataset
As, Xs, y = datagen('dblp')

# Initialize and fit LMGEC model
model = LMGEC(beta=2, temperature=10)
labels = model.fit_predict(X)

# Evaluate clustering
from sklearn.metrics import adjusted_rand_score
print(f"ARI: {adjusted_rand_score(y, labels):.3f}")

Documentation

Full documentation is available at: https://gackouhamady.github.io/mvcluster/

** Includes:

• API reference

• Tutorials

• Theory behind the algorithms

• Contribution guidelines

Research Context

Developed during my research internship at Centre Borelli under the supervision of Lazhar Labiod. The project focused on:

• Developing efficient algorithms for multiview graph clustering

• Creating a user-friendly Python package

• Benchmarking against existing methods

• Applying techniques to real-world datasets

Supported Datasets

• DBLP academic network

• ACM citation network

• IMDB movie database

• Amazon product graphs

• Wikipedia article network

• Aloi

• Mfeat

• Arabidopsis

Contributing

We welcome contributions! Please see our Contribution Guidelines for details.

Citation

If you use mvcluster in your research, please cite:

@inproceedings{fettal2023efficient, author = {Fettal, Chakib and Labiod, Lazhar and Nadif, Mohamed}, title = {Simultaneous Linear Multi-view Attributed Graph Representation Learning and Clustering}, booktitle = {Proceedings of the 16th ACM International Conference on Web Search and Data Mining}, year = {2023}, doi = {10.1145/3539597.3570367} }

Contact

For questions or support, please contact: [[e-mail]]

License: MIT Status: Active development Python versions: 3.8+ Source Code: https://github.com/gackouhamady/mvcluster