search

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

Reproduce the results on wadi #9

Closed ccskyyy closed 2 years ago

ccskyyy commented 3 years ago

Your work is amazing and impressive. When I was trying to reproduce your results on the wadi dataset, there were some problems that I was not clear about.

  1. I used WADI_14days_new.csv as training data, WADI_attackdataLABLE.csv as a test one. Did you also train on those files?
  2. I saw there are four empty columns in the wadi dataset('2_P_001_STATUS', '2_P_002_STATUS', '2_LS_001_AL','2_LS_002_AL'), how did you process these columns? (Did you just drop these four empty columns?)
  3. there are some nan values in the dataset, how did you deal with them in the paper? (I replaced these values with the mean of the columns'.)

And I got these results on wadi data: F1 score: 0.3994959042218021 precision: 0.5382003395585738 recall: 0.31795386158475425

So I was wondering how did you process the dataset? Any suggestions would be appreciated! Thank you so much! :)

d-ailin commented 3 years ago
  1. No, we only train on the former data.
  2. We would fill them as zeros.
  3. Yes, we also replace them with the mean of the columns.

We use min-max normalization and you could also check with the detail(e.g. downsampling) in the paper. Thanks.

Notes: please do the min-max normalization first and then downsample.