knowledgedefinednetworking / demo-routenet Goto Github PK

Hello,
it seems like that I don't understand how to predict jitter and packet loss ratio,I just have tried to replace “delay” with "jitter" in the definition of "target",and I didn't modify anywhere else,the result is too strange to tolerate. So I think there's propably something wrong .

If I want to predict jitter and packet loss ratio with your jupyter notebook,where should I modify?Could you please just tell me how to do?
Thank you very much.

For what it's worth, I ran the tf_upgrade_v2 script on the code and, aside from a very simple change to HParams, I was able to get this to run in the latest TF2.x. Not sure if you are still maintaining, but it's about a 5 minute fix.

Hi, i have successfully run this demo in Ubuntu, but when i ran it in pycharm(2019.1 version) in windows 10, it cannot start training, i attached the information in pycharm terminal in the fowllowing .txt file. The NVIDIA GNU TOOL KIT(CUDA+CUDNN) have been installed， please have a look at the attached .txt file, thanks.
demoroutenet_pycharm.TXT

Hi~I'm a student who is interested in your work. I also want to use your datasets on my own network to predict the performance of network. I see you replied that you may open the simulator code and make it public. I want to know if that has been done. If did, I'll be grateful that you tell me the site. If not, thanks for your work, too.

Hello, I'm a student who's been working with your paper "Unveiling the potential of GNN~". I've done some experiments with the datasets you uploaded, but I want to know how those datasets are made. I want to test your code to my own network but I can't get hold of how it was made. It seems that omnetpp does not support your dataset type. So it'd be great if you also upload how you made your datasets.

Hello,
now I can use your source code and datasets to get accurate predictions, but I want to know how can I use the model to predict with self-defined configurations?Or specificly, how can I build my own datasets?
Thank you very much.

Are the software versions fixed for this challenge? I can not install it on Manjaro 5.7.0-3.

ERROR: Could not find a version that satisfies the requirement tensorflow-gpu==1.14.0 (from -r requirements.txt (line 20)) (from versions: 2.2.0rc1, 2.2.0rc2, 2.2.0rc3, 2.2.0rc4, 2.2.0, 2.3.0rc0)
ERROR: No matching distribution found for tensorflow-gpu==1.14.0 (from -r requirements.txt (line 20))