This is a repository for Data Mining Mid-term Homework in SJTU

The core idea of EGAE is to design a GNN to find an ideal space for the relaxed k-means on graph data. We prove that the relaxed k-means will obtain a precise clustering result under some strong assumptions. So we attempt to use GNNs to map the data into an ideal space that satisfies the strong assumptions.

python run.py

pytorch >= 1.3.1

scipy 1.3.1

scikit-learn 0.21.3

numpy 1.16.5

• model.py: An efficient implementation which can be used when datasets are not too large.
• sparse_model.py: It is a sparse implementation of EGAE for large scale datasets, e.g., PubMed.