search

d-ailin / GDN

Implementation code for the paper "Graph Neural Network-Based Anomaly Detection in Multivariate Time Series" (AAAI 2021)
MIT License
490 stars 142 forks source link

The model does not perform well #80

Open mhy139 opened 1 year ago

mhy139 commented 1 year ago

Hello author, thank you very much for your work.When I run your code, after using the wadi data set in your paper and the parameters you required, the F accuracy is particularly low, how can I solve it

d-ailin commented 1 year ago

Thanks for your interest. Based on the previous issues, one major reason is the preprocessing, which you could access via https://github.com/d-ailin/GDN/tree/main/scripts. You could also see some previous issues such as #52 for more details.

mhy139 commented 1 year ago

Thank you for your answer. The A2 version of my data set is used. After I downloaded the A1 version, I found that there was no marked anomaly in the WADI_attackdata.csv file. If possible, you can send a copy of your data set to the mailbox mahongye77@gmail.com