Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
You can not select more than 25 topics
Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
Hongyi Zhang
925b86798b
|
1 year ago | |
|---|---|---|
| dgl_example | 1 year ago | |
| openhgnn_example | 1 year ago | |
| pyg_example | 1 year ago | |
| .gitignore | 1 year ago | |
| README.md | 1 year ago | |
README.md
icdm 2022 baseline
We implement RCCN baseline on ICDM 2022 : Risk Commodities Detection on Large-Scale E-commerce Graphs. Our implementation include three popular GNN platforms: DGL, PyG and OpenHGNN. You can use it as a reference for data processing, model training and inference.
Environment Setup
- Option1: Run on OpenI Cloud platform.
- We provide an image with pytorch, dgl and pyg installed. You can directly create an instance from that. Also, you can use dataset provided on OpenI.
- Option2: Run locally.
- Download the dataset and config the environment locally.
Experiment Result
| AP | session1 | session2 |
|---|---|---|
| pyg | 0.9174 | 0.8931 |
| dgl | 0.8751 | 0.8417 |
| openhgnn | 0.8658 | 0.8387 |