Awesome

SCAN: Semi-supervisedly Co-embedding Attributed Networks

This repository contains the Python&Pytorch implementation for SCAN. Further details about SCAN can be found in this paper:

Zaiqiao Meng, Shangsong Liang, Jinyuan Fang, Teng Xiao. Semi-supervisedly Co-embedding Attributed Networks. (NeurIPS 2019)

The orignal tensorflow implementation for SCAN can be found in SCAN

Requirements

=================

• Pytorch (1.0 or later)
• python 3.6/3.7
• scikit-learn
• scipy

Run the demo

=================

python train.py

Result

The Link prediction performance AUC&AP score :

The Attribute inference performance AUC&AP score :

The node classification performance accuracy :