issues about the metis algorithm

benedekrozemberczki / ClusterGCN

A PyTorch implementation of "Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks" (KDD 2019).

GNU General Public License v3.0

786 stars 135 forks source link

issues about the metis algorithm #31

Open immortal13 opened 3 years ago

immortal13 commented 3 years ago

(st, parts) = metis.part_graph(self.graph, self.args.cluster_number) Thanks for your awesome code, could you please tell me how metis conduct the graph partition? Cause the self.graph here doesn't include the information about edge weights and feature attributes.

Orion-wyc commented 3 years ago

Perhaps, you can find the answer in METIS. Metis only partitions the nodes into parts ( the example in pymetis can help you better understand it), which means you need to split the features and edge weights manually.

If you know DGL, the dgl.node_subgraph function can help split the features.

immortal13 commented 3 years ago

thank you very much! It is very helpful.

qwertyuiopqwe123 commented 8 months ago

How to generate the three files in this ’input‘ when i wanna run the new dataset？