zero false positive and negative

[abdullahasif07]

Hi..im getting zero false positives and negatives for s1-s4 experiments. What could be a reason for this as i have looked carefully into it and am not able to figure out. Secondly there are multiple malicious_labels.txt files in training logs of s1,s2,s3,s4 so are we supposed to run training keeping all of them or the most recent copy?

[paxl0807]

I have encountered the same problem as yours. According to the paper, the "malicious_labels.txt" file of the S-1 dataset should contain 22 labels, but in the actual code, only 3 labels are provided. Have you finally solved this problem?