GIN
========================
- Paper link: https://arxiv.org/abs/1810.00826
- Author's code repo (in Pytorch): https://github.com/weihua916/powerful-gnns
Run with following:
# use tensorflow bakcend
TL_BACKEND=tensorflow python gin_trainer.py --dataset=MUTAG # use pytorch backend
TL_BACKEND=torch python gin_trainer.py --dataset=MUTAG # use paddle backend
TL_BACKEND=paddle python gin_trainer.py --dataset=MUTAG
| Dataset | Paper | Our(pd) | Our(tf) | Our(th) |
|---|---|---|---|---|
| MUTAG | 89.4 ± 5.6 | 89.4 ± 5.6 | 89.4 ± 5.6 | 89.4 ± 5.6 |