hkuds/stexplainer

[CIKM'2023] "STExplainer: Explainable Spatio-Temporal Graph Neural Networks"

Hi there 👋

✨Welcome to the Data Intelligence Lab @ HKU!✨

🚀 Our Lab is Passionately Dedicated to Exploring the Forefront of the Data Science & AI 👨‍💻

       

stexplainer's Issues

Data & Explainer

I hope you have a good day.

You did a good job in STExplainer and we are trying to follow your work. However, I'm facing two problems:

1. Where should I find other datasets besides PEMS04 formatted for the experiments?
2. Could you please provide the code for explaining the procedure? Here is only the code for training/testing STExplainer on the regression task. We are looking for the code for the explaining task, eg: generating and evaluating the explanation sub-graphs.

I'm looking forward to hearing from you.

Best regards

The methods/pytorch_net folder is missing files

The methods/STExplainer/pytorch_net folder is missing files, can you add them? Thank you!

Data Missing Experimental Setting

Hello, very great works!
Could you provide more details and codes for how to conduct data missing robustness experiments?
Thank you for consideration!

How to conduct the Crime Data (NYC, CHI) experiment

Now we can only reproduce the experiment of PEMS4 data set, how to analyze the crime data?
If convenient, could you tell me how to obtain the raw data (NYC, CHI) and conduct the experiment using your method.
In addition, whether your project must run in Ubuntu environment. It is difficult to configure dependency packages in Windows 10.