Graph Pre-Training Models Are Strong Anomaly Detectors

Quick Start

To reproduce the reported results, please run the script with --use_cfg.

Node-level Graph Anomaly Detection

# With best configurations
python pretrain_node.py --dataset <dataset> --use_cfg

Supported datasets includes weibo, reddit, amazon, yelp, tfinance, elliptic, tolokers, questions. dgraphfin, tsocial

Graph-level Graph Anomaly Detection

# With best configurations
python pretrain_graph.py --dataset <dataset> --use_cfg

Supported datasets includes DD, IMDB-BINARY, REDDIT-BINARY, PROTEINS_full, AIDS, NCI1, Mutagenicity

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
config		config
README.md		README.md
dataset.py		dataset.py
evaluation.py		evaluation.py
model.py		model.py
pretrain_graph.py		pretrain_graph.py
pretrain_node.py		pretrain_node.py
utils.py		utils.py