Why don't you normalize the input data?

d-ailin / GDN

Implementation code for the paper "Graph Neural Network-Based Anomaly Detection in Multivariate Time Series" (AAAI 2021)

MIT License

482 stars 141 forks source link

Why don't you normalize the input data? #23

Closed Scienceseb closed 2 years ago

Scienceseb commented 2 years ago

According to your code in TimeDataset.py you import a Scikit-learn module to do the normalization of the data, but it is never used. Don't you think it could be harmful for the model if there is a big difference in scale between the different measurements?

Thanks !

Kind regards, Sébastien de Blois

d-ailin commented 2 years ago

As the normalization has been moved to the preprocessing part, you could refer to the https://github.com/d-ailin/GDN/tree/main/scripts for the detail of preprocessing.