21 Feb
2023
21 Feb
'23
8:49 a.m.
Dear Xiaolan Xie,
thank you for your interest in our work. The dimension 32 from the
link_inputs comes from the link hidden state size. The other two dimensions
are not fixed, that's why it appears with the symbol "?". This is because
there can be different number of paths and links for each topology.
Therefore, with "?" it indicates that the dimension is variable. For more
details about the shape of link_inputs see
https://github.com/knowledgedefinednetworking/demo-routenet/blob/6fe560081f….
Best,
Paul Almasan
On Mon, 20 Feb 2023 at 07:51, 谢小兰 <21032303148(a)mails.guet.edu.cn> wrote:
Dear Pro J. Suárez-Varela, S. Carol-Bosch, K. Rusek,
P. Almasan, M.
Arias, P. Barlet-Ros, A. Cabellos-Aparicio:
Hello Pro J. Suárez-Varela, S. Carol-Bosch, K. Rusek, P. Almasan, M.
Arias, P. Barlet-Ros, A. Cabellos-Aparicio, I am sorry to disturb you in
your busy schedule. I'am a student of Guilin University of Electronic
Science and Technology from China.
Recently, Iam learning how to do route optimization with grahp neural
networks. Your paper RouteNet: Leveraging Graph Neural Networks for network
modeling and optimization in SDN gives me great inspiration. I ran the
model code after reading your paper, but I can't see how much about the
model regarding the input and output dimensions, for example, the shape of
link_inputs=(? ,? ,32). I would like to ask about this? What is the size of
the How can I can see it in the code?
No matter whether you help me or not, best wishes to you. I would
appreciate it if you could help me.
Thanks and regards.
Sincerely,
Xiaolan, Xie
2023/2/19
_______________________________________________
Kdn-users mailing list
Kdn-users(a)knowledgedefinednetworking.org
https://mail.n3cat.upc.edu/cgi-bin/mailman/listinfo/kdn-users