Paper: Heterogeneous Graph Propagation Network
The author did not provide codes. So, we complete it according to the implementation of HAN
Clone the Openhgnn-DGL
python main.py -m HPN -t node_classification -d acm_han_raw -g 0If you do not have gpu, set -gpu -1.
Candidate dataset: acm_han_raw, acm4GTN, imdb4GTN
| HPN[OpenHGNN] | Macro-F1 | Micro-F1 |
|---|---|---|
| acm_han_raw | 91.80 | 91.80 |
| acm4GTN | 91.04 | 90.92 |
| imdb4GTN | 60.96 | 64.00 |
The model is trained in semi-supervisied node classification.
HPNLayer
SemanticFusion
Supported dataset: acm_han_raw, acm4GTN, imdb4GTN
You can download the dataset by
wget https://s3.cn-north-1.amazonaws.com.cn/dgl-data/dataset/acm_han_raw.zip
wget https://s3.cn-north-1.amazonaws.com.cn/dgl-data/dataset/acm4GTN.zip
wget https://s3.cn-north-1.amazonaws.com.cn/dgl-data/dataset/imdb4GTN.zip
k_layer = 1 # iterations in Semantic Propagation
alpha = 0.1 # Value of restart probability
out_embedsize = 64 # Dimension of the final embedding
edge_drop = 0 # the dropout rate on edges that controls the messages received by each nodeBest config can be found in best_config
Donglin Xia[GAMMA LAB]
Submit an issue or email to donglin.xia@bupt.edu.cn.