About the results in MSL dataset

[LukaZLionel]

Nice work! But I met a problem, when I run the script in your code to MSL dataset, one of the results F-score is only 0.9397, which is not same as 0.96 in the report.
I wonder if there is something I missed.

[tianzhou2011]

A quick run: This set of parameters yields an F1 score of 95.5 for me. However, Yiyuan may employed different sets of parameters resulting in an F1 score of 96.6. I apologize for any inconvenience caused, and please understand that due to busy schedule, it may take some time for him to reply.

[yyysjz1997]

Nice work! But I met a problem, when I run the script in your code to MSL dataset, one of the results F-score is only 0.9397, which is not same as 0.96 in the report. I wonder if there is something I missed.

Thank you for your feedback. In the paper, we presented the results after averaging, but the final outcomes might be influenced by hardware conditions, the python package version and shuffling operation, among other factors. Further testing can be conducted by altering certain hyperparameters, such as window size, patch size and so on. I hope this helps.