dmlc--dgl
Lingfan Yu
74e13eea61
* fix gat code to use latest edge softmax module * avoid transpose * update README * use edge_softmax op * mxnet edge softmax op * mxnet gat * update README * fix unittest * fix ci * fix mxnet nn test; relax criteria for prod reducer
24 行
547 B
Markdown
24 行
547 B
Markdown
Graph Attention Networks (GAT)
|
|
============
|
|
|
|
- Paper link: [https://arxiv.org/abs/1710.10903](https://arxiv.org/abs/1710.10903)
|
|
- Author's code repo:
|
|
[https://github.com/PetarV-/GAT](https://github.com/PetarV-/GAT).
|
|
|
|
Note that the original code is implemented with Tensorflow for the paper.
|
|
|
|
### Dependencies
|
|
* MXNet nightly build
|
|
* requests
|
|
|
|
```bash
|
|
pip install mxnet --pre
|
|
pip install requests
|
|
```
|
|
|
|
|
|
### Usage (make sure that DGLBACKEND is changed into mxnet)
|
|
```bash
|
|
DGLBACKEND=mxnet python3 train.py --dataset cora --gpu 0 --num-heads 8
|
|
```
|