Reputation: 4290
well i am trying to use community detection algorithms by networkx on famous facebook snap data set. here are my codes :
import networkx as nx
import matplotlib.pyplot as plt
from networkx.algorithms import community
from networkx.algorithms.community.centrality import girvan_newman
G_fb = nx.read_edgelist("./facebook_combined.txt",create_using = nx.Graph(), nodetype=int)
parts = community.best_partition(G_fb)
values = [parts.get(node) for node in G_fb.nodes()]
but when i'm run the cell i face with the title error which is :
AttributeError: module 'networkx.algorithms.community' has no attribute 'best_partition'
any advice ?
Upvotes: 16
Views: 30636
Reputation: 303
This has helped me to run the code without errors:
pip uninstall community
import community.community_louvain as cl
partition = cl.best_partition(G_fb)
Upvotes: 0
Reputation: 1988
I naively thought that pip install community
was the package I was looking for but rather I needed pip install python-louvain
which is then imported as import community
.
Upvotes: 4
Reputation: 11
I also faced this in CS224W but changing the karate.py or other solutions didn't work.
For me (in colab) using the new PyG installation code worked. this code, will install the last version:
!pip install -q torch-scatter -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
!pip install -q torch-sparse -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
!pip install -q git+https://github.com/rusty1s/pytorch_geometric.git
Upvotes: 1
Reputation: 658
I faced this in CS224W
AttributeError: module 'community' has no attribute 'best_partition'
Pls change this file karate.py
replace import to
import community.community_louvain as community_louvain
then it works for me.
Upvotes: 13
Reputation: 41
I had the same problem. In my case, it was solved importing the module in a different manner:
import community.community_louvain
Upvotes: 4
Reputation: 731
I had a similar issue. In my case, it was because on the other machine the library networkx was obsolete.
With the following command, the issues was solved.
pip3 install --upgrade networkx
Upvotes: 0
Reputation: 353419
I think you're confusing the community module in networkx proper with the community detection in the python-louvain module which uses networkx.
If you install python-louvain, the example in its docs works for me, and generates images like
Note that you'll be importing community
, not networkx.algorithms.community
. That is,
import community
[.. code ..]
partition = community.best_partition(G_fb)
Upvotes: 13