the kdd dataset is different from other kdd datasets, and how was it processed？

[Irisansan]

Hi，
I have noticed that there is only 35 features in the kdd dataset given in this project，which is different from other kdd datasets including official ones. And there is no description for how it has been processed，so I am confused.
Could anyone give me a hand？I will be grateful！
Regards,
Iris

[GPla]

The data comes from here, you can find the list of features under task description. As you mentioned, there are 35 features listed (label not listed on the page). If we compare them with other sources, such as datahub, we can see that they use 7 additional features with information about the destination host. So no preprocessing was done, there are just different version available.

[Irisansan]

Got it！Thank you very much！！！

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "GPla"<[email protected]&gt;; 发送时间: 2020年4月30日(星期四) 晚上8:31 收件人: "LiDan456/MAD-GANs"<[email protected]&gt;; 抄送: "甘樂"<[email protected]&gt;; "Author"<[email protected]&gt;; 主题: Re: [LiDan456/MAD-GANs] the kdd dataset is different from other kdd datasets, and how was it processed？ (#13) The data comes from here, you can find the list of features under task description. As you mentioned, there are 35 features listed (label not listed on the page). If we compare them with other sources, such as datahub, we can see that they use 7 additional features with information about the destination host. So no preprocessing was done, there are just different version available. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[Irisansan]

The data comes from here, you can find the list of features under task description. As you mentioned, there are 35 features listed (label not listed on the page). If we compare them with other sources, such as datahub, we can see that they use 7 additional features with information about the destination host. So no preprocessing was done, there are just different version available.

Hey, Have you got the satisfied performance values on SWaT data？Can you tell me how to adjust the parameters settings？Thanks a lot.

[huaxinfeihua]

The data comes from here, you can find the list of features under task description. As you mentioned, there are 35 features listed (label not listed on the page). If we compare them with other sources, such as datahub, we can see that they use 7 additional features with information about the destination host. So no preprocessing was done, there are just different version available.

The dataset from http://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html has 42 features and no labels, but the code it is have 35 and get labels,What do I have to do to get these labels, please?

[luckyStarForever]

Would you like to tell me , "What are the requirements for the code project environment ?"

You can also run commands in the project environment:

pip freeze > D:\requirements.txt

Please share the requirements.txt content.
Thank you very much!