https://github.com/hiroakimikami/nnabla-ggnn

Reimplement gated graph neural network in NNabla
https://github.com/hiroakimikami/nnabla-ggnn

deep-learning deep-neural-networks ggnn nnabla

Last synced: 28 days ago
JSON representation

Reimplement gated graph neural network in NNabla

• Host: GitHub
• URL: https://github.com/hiroakimikami/nnabla-ggnn
• Owner: HiroakiMikami
• Created: 2018-04-15T10:06:04.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-04-30T10:44:24.000Z (almost 7 years ago)
• Last Synced: 2024-11-13T17:52:23.385Z (3 months ago)
• Topics: deep-learning, deep-neural-networks, ggnn, nnabla
• Language: Python
• Size: 19.5 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
nnabla-ggnn: NNabla Implementation of GG-NN

===

This repository is a NNabla implementation of Gated Graph Sequence Neural Networks (GG-NN) proposed in the paper [Gated Graph Sequence Neural Networks](https://arxiv.org/abs/1511.05493) by Y.Li, D.Tarlwo, M.Brockschmdit, and R. Zemel. GG-NNs can use graph-structured data as inputs of neural networks, and gets high accuracy on some bAbI-tasks. This implementation is tested with bAbI 15 and bAbI 19, and gets high accuracy (100% for bAbI15 and 95% for bAbI 19).

The official implementation is available in the [GitHub repository](https://github.com/yujiali/ggnn).

Requirements

---

* Python 3.x (tested with Python 3.6.5)

* NNabla 0.9.9

Run Examples

---

### bAbI 15

```bash

$ babi-tasks 15 1000 > train.txt # Notes: babi-tasks can be installed from https://github.com/facebook/bAbI-tasks

$ babi-tasks 15 1000 > vaild.txt

$ python ./main.py bAbI15 --train-file train.txt --valid-file valid.txt # --context cudnn

```

My result: get 100% validation accuracy after 200 iterations (1 epochs).

### bAbI 19

```bash

$ babi-tasks 19 1000 > train.txt # Notes: babi-tasks can be installed from https://github.com/facebook/bAbI-tasks

$ babi-tasks 19 1000 > vaild.txt

$ python ./main.py bAbI10 --train-file train.txt --valid-file valid.txt # --context cudnn

```

My result: get 95% validation accuracy after 54000 iterations (216 epochs).

TODO

---

* [ ] mini-batched training

    * I didn't implement the mini-batch version.

    * I think it is not difficult to implement mini-batched training if the number of vertices in graph is same.

References

---

* [Gated Graph Sequence Neural Networks, ICLR 2016](https://arxiv.org/abs/1511.05493)

* [the official GitHub repository](https://github.com/yujiali/ggnn)

* [bAbI-tasks](https://github.com/facebook/bAbI-tasks)