Closed Tim-Cunnane closed 5 months ago
I see, you failed to produce the poisoning dataset of the GAN-based dataset. First, please make sure you get the latest version of code. Check again if the ctgan_samples.csv
located in the folder /maip/src/server/deep-learning/attacks/model_2
on your machine. Then, run again the server. If you perform successfully the GAN attack, you should see two new csv on the same folder: ctgan_poisoned_dataset.csv
and ctgan_poisoned_dataset_view.csv
(for visualization purpose only).
Hope that helps!
Hi,
Ahh got it sorted thank you for you help. How long does it usually take to get the resilience metrics for before and after the attack.
Just to let you know the node modules folder isn't on the repo anymore which stop me from getting on the the GUI
Thanks, Tim
Hi,
Sorry to be a hassle, How long does it usually take to get the resilience metrics for before and after the attack? Getting the sort of thing with the logs ![Uploading image.jpg…]()
Thanks, Tim
The entire process (producing a poisoned GAN dataset and computing resilience metrics) took around 10-15 seconds. The second step involves retraining the model using the poisoned dataset generated from the first step. Both steps depend on the size of the GAN dataset; however, in this case, it is quite fast.
I have been running the resilience for quite a while now would there be any specific reason why it would be taking longer to complete.
Thanks
@Tim-Cunnane: Can you please provide a copy of the server log when you compute the resilience metrics? It is difficult for me to debug without it. If you want compute resilience metrics for other models (not model_2
), you may need to copy the GAN dataset ctgan_samples.csv
to the corresponding folder of the tested model in /attacks
.
The server is below:
I also receive this error sometimes when start the resilience metric:
@Tim-Cunnane: It's a bit strange as I have not encountered this issue. Please check again if the poisoned GAN dataset is successfully generated as below. Also, calculating the resilience metrics requires retraining the model. I'm not sure if the problem is due to your hardware issues.
Hi,
Yeah all of the poisoned dataset are there within that folder. I think i may have found the issue where it is the AVX instruction but i have install the link that you suggest in the above comment and i am getting this when running the client of MAIP would there be any specific reason for this? The repo is fully up-to-date
Thanks
Stange why the above is happen and i am stuggling to figureout the reason for this.
Thanks
Sorry Tim, unfortunately I don't have a solution for the hardware issue as it worked totally fine for me.
That is not a bother, just wanted to check what type of AVX instruction does the CPU need to be compatible for such as AVX512, AVX1 or 2?
Hi,
Sorry i believe that i have found a laptop that will allow me to be able complete the tasks. I came across an issue when installing the nodejs using this command curl -sL https://deb.nodesource.com/setup_19.x | bash
I am receiving this error:
I was wondering if you would know a fix to this issue?
Appreciate any help that you give me.
Thanks
Nice try. You can copy the error message on google, I found this potential solution.
... | sudo bash
Hope this works.
Hi,
Thank you for all of your help, Just want to see if you had any other pcaps from the public dataset that you got the botnet pcap from such as the DDos attack or DoS, etc.
Thanks
Hi, I am currently in the process of setting up the maip platform on the VM, however I am having a couple of issues with this. I have the UI started but if I go into any section of the system it will say error{} in the top right corner of the screen. Would you be able to give me a hand with this?
Thanks