How can I use my own data set for embedding operations?

divelab / lgcn

GNU General Public License v3.0

136 stars 65 forks source link

How can I use my own data set for embedding operations? #9

Open rwN8Z8n3 opened 5 years ago

rwN8Z8n3 commented 5 years ago

Thanks for your wonderful contribution. But I don't see how to build a dataset in the form of sample you give. Can you tell me how I should build my dataset for embedding?

HongyangGao commented 5 years ago

Hi, it depends on your raw data. The key idea is to produce node features for each node in graph and generate the adjacency matrix to use.

rwN8Z8n3 commented 5 years ago

Thanks for your help! However, in your code (utils.py), there are seven small data files (like ['x', 'y', 'tx', 'ty', 'allx', 'ally', 'graph']) as input, how should I preprocess my network files to the corresponding format?