A PyTorch Implementation of GGNN

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 gets 100% accuracy on node-selection bAbI task 4, 15, and 16. Their official implementation are available in the yujiali/ggnn repo on GitHub.

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==2.7

• PyTorch>=0.2

Run

Train and test the GGNN:

python main.py --cuda (use GPUs or not)

Suggesting configurations for each task:

# task 4
python main.py --task_id 4 --state_dim 4 --niter 10
# task 15
python main.py --task_id 15 --state_dim 5 --niter 10
# task 16
python main.py --task_id 16 --state_dim 10 --niter 150

Results

I followed the paper, randomly picking only 50 training examples for training. Performances are evaluated on 50 random validation examples.

| bAbI Task | Performance | | ------| ------ | | 4 | 100% | | 15 | 100% | | 16 | 100% |

Here's an example of bAbI deduction task (task 15)

Disclaimer

The data processing codes are from official implementation yujiali/ggnn.

TODO

• [ ] GraphLevel Output

References

• Gated Graph Sequence Neural Networks, ICLR 2016

• yujiali/ggnn