louvain

0.8.2 verified Fri May 01 auth: no python

louvain is a Python implementation of the Louvain algorithm for community detection in large networks, based on the igraph library. Current version is 0.8.2, with a relatively stable release cadence (last update 2024).

pip install louvain

Common errors

error ModuleNotFoundError: No module named 'igraph' ↓

cause louvain requires igraph but doesn't install it as a dependency

fix

pip install python-igraph

error ImportError: cannot import name 'find_partition' from 'louvain' ↓

cause Old API (v0.5.x) used 'best_partition'; newer versions use 'find_partition'

fix

Use: partition = louvain.find_partition(G, louvain.ModularityVertexPartition)

error AttributeError: module 'louvain' has no attribute 'best_partition' ↓

cause The 'best_partition' function was removed after v0.5.x

fix

Use louvain.find_partition() with appropriate partition class

Warnings

deprecated The standalone 'louvain' package is deprecated; consider using 'leidenalg' for newer algorithms ↓

fix Use pip install leidenalg and import leidenalg (also by same author, more efficient)

gotcha Must install python-igraph separately ↓

fix Run: pip install python-igraph

gotcha The 'louvain' module does not install igraph automatically; missing igraph causes ImportError ↓

fix Ensure igraph is installed and imported before louvain

Imports

louvain
```
import louvain
```

Quickstart

Basic community detection using the Louvain algorithm on a random graph.

import igraph as ig
import louvain

G = ig.Graph.Erdos_Renyi(n=100, p=0.1, directed=False)
partition = louvain.find_partition(G, louvain.ModularityVertexPartition)
print(partition.membership)