Skip to content

udayshankars/GrAMME

master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
lib
 
 
 
 
 
 

GrAMME

GrAMME: Semi-Supervised Learning using Multi-layered Graph Attention Models

In this work, we perform semi-supervised learning on multi-layer graphs using attention models. We have proposed two attention models for effective feature learning, GrAMME-SG and GrAMME-Fusion, that exploit the inter-layer dependencies for building multi-layered graph embeddings. Both are architectures are depicted in the figures below:

GrAMME-Fusion

GrAMME-Fusion

GrAMME-SupraGraph

GrAMME-SupraGraph

Usage


Example Usage python run_gramme_fusion.py --multiplex_edges_filename multilayer.edges --multiplex_labels_filename multilayer.labels

multilayer.edges file should be an edgelist along with layer information as shown below:

layer_ID node_ID node_ID

layer_ID starts from 0 and node_ID starts from 0

--multiplex_edges_filename: multilayer.edges, eg::

0 0 1
0 1 0
0 2 5
0 5 2
...
1 0 3
1 3 0
1 4 9
1 9 4
...
2 0 7
2 7 0
2 1 3
2 3 1
...

Note: For undirected edges between node u and v, edgelist should contain both the entries u-> v and v-> u

0 u v
0 v u

multilayer.labels should only contain the true labels of the nodes (explicit node ids should not be there), class labels start from 0 to C-1 where C is the total number of classes

multiplex_labels_filename: multilayer.labels, eg::

0
0
2
1
9
0
3
...

Requirements

  • python 3.6
  • numpy >= 1.11.0
  • tensorflow >= 1.5.0
  • scikit-learn >= 0.18
  • matplotlib >= 2.1.0

Citations

If you find GrAMME useful in your research, please cite the following paper:

@article{shanthamallu2018attention,
  title={GrAMME: Semi-Supervised Learning using Multi-layered Graph Attention Models},
  author={Shanthamallu, Uday Shankar and Thiagarajan, Jayaraman J and Song, Huan and Spanias, Andreas},
  journal={arXiv preprint arXiv:1810.01405},
  year={2018}
}

About

GrAMME: Semi-Supervised Learning using Multi-layered Graph Attention Models

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages