Skip to content

[UNMAINTAINED] A PyTorch Implementation of Gated Graph Sequence Neural Networks (GGNN) for Graph Classification

License

Notifications You must be signed in to change notification settings

bdqnghi/ggnn_graph_classification

 
 

Repository files navigation

A PyTorch Implementation of GGNN for Graph Classification

This is a PyTorch implementation of the Gated Graph Sequence Neural Networks (GGNN) as described in the paper Gated Graph Sequence Neural Networks by Y. Li, D. Tarlow, M. Brockschmidt, and R. Zemel.

This implementation focuses on the Graph Level output, which hasn't been exploiting from the original code base. In concrete, we focus the Graph Classification task, which requires the Graph Level output to be implemented.

We took the dataset of 104 programming problems, which comprises of 52000 cpp files from the paper Convolutional Neural Networks over Tree Structures for Programming Language Processing and parse the cpp files into the graph representation based on the details of the paper Learning to Represent Programs with Graphs.

What is GGNN?

  • Solve graph-structured data and problems
  • A gated propagation model to compute node representations
  • Unroll recurrence for a fixed number of steps and use backpropogation through time
  • An output model to make predictions on nodes

Requirements

  • python==3.6
  • PyTorch>=0.2

Run

Train and test the GGNN:

python3 main_ggnn.py --training --cuda (use GPUs or not)

References

About

[UNMAINTAINED] A PyTorch Implementation of Gated Graph Sequence Neural Networks (GGNN) for Graph Classification

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Java 90.1%
  • Python 8.0%
  • Shell 1.9%