zhumeiqiBUPT / AM-GCN

AM-GCN: Adaptive Multi-channel Graph Convolutional Networks
MIT License
219 stars 46 forks source link

use model with different data set #17

Open Mopsfrosch opened 3 years ago

Mopsfrosch commented 3 years ago

Hey! I would like to use your model with a different data set and therefore need to first produce the correct input files. Can you give some more details on how to do this? in the file dataprocess.py for example def process_data(dataset): names = ['y', 'ty', 'ally','x', 'tx', 'allx','graph'] there are files expected in "../data/cache/ind.{}.{}".format(dataset, names[i]),.. for each in names. What are these files? They aren't provided in the repository (the cache folders are missing)

I would be very grateful if you could provide the files so that I can replicate the processing, or provide an explanation on how to create the input for a new data set.

Thanks in Advance! Johanna

Mopsfrosch commented 3 years ago

additional question: is it possible to use the model with multi-label tasks?