search

SherylHYX / SSSNET_Signed_Clustering

Official code for the SDM2022 paper -- SSSNET: Semi-Supervised Signed Network Clustering.
MIT License
22 stars 2 forks source link

SPONGE rainfall labels #5

Open dinhtrang24 opened 1 week ago

dinhtrang24 commented 1 week ago

Hi, Thank you for the excellent work! I would like to use the rainfall dataset with SPONGE labels for a node classification task. Could you please direct me to where I can find the labels generated by SPONGE that categorize the 306 locations into 6 clusters corresponding to geographic locations? I initially thought these labels were stored in the file labels6SPONGE.npy, but when I plotted them against the locations, they didn’t seem to align, as shown in Figure 10 of the paper "SPONGE: A generalized eigenproblem for clustering signed networks." Could you clarify this for me? Thank you!

SherylHYX commented 1 week ago

I am sorry about the confusion but the labels were generated by using the signet package and 

Aps = ss.load_npz('plus_cc.npz')

Ams = ss.load_npz('minus_cc.npz')

m = Cluster((Aps, Ams))

k = 6 # number of clusters

labels = m.SPONGE(k)

If you notice any subtle misuse of the signet package from me, please let me know, and I would be more than happy to correct this. Thank you for your interest and your rigorous check on this issue!