Doing graph-level classification?

[gailmargolis76]

My input is such that each subject has their own graph. This is different from the example given in train.py where there is only 1 graph (a citation network). In the tensorflow implementation of gcn, you suggest doing graph-level classification by combining the adjacency matrices of all the graphs in the input sample into one large adjacency matrix (as a sparse block-diagonal matrix). The part I am not sure how to implement in keras is the pooling of the output to produce 1 classification per graph. Any tips would be greatly appreciated!

[tkipf]

Have a look here: tkipf/gcn#4. This only applies for the TensorFlow implementation though. Hope this helps!

…

On Wed 27. Jun 2018 at 23:42 gailmargolis76 ***@***.***> wrote: My input is such that each subject has their own graph. This is different from the example given in train.py where there is only 1 graph (a citation network). In the tensorflow implementation of gcn, you suggest doing graph-level classification by combining the adjacency matrices of all the graphs in the input sample into one large adjacency matrix (as a sparse block-diagonal matrix). The part I am not sure how to implement in keras is the pooling of the output to produce 1 classification per graph. Any tips would be greatly appreciated! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#33>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AHAcYJWmmhWbZ80CLdpev3L8aMxNscgTks5uBApogaJpZM4U6flP> .

[tkipf]

Ah, i just realized you’re explicitly asking about how to do this in keras. This is a bit tricky due to the restrictions of the API and I would recommend going with pure TF in this case.

…

On Wed 27. Jun 2018 at 23:46 Thomas Kipf ***@***.***> wrote: Have a look here: tkipf/gcn#4. This only applies for the TensorFlow implementation though. Hope this helps! On Wed 27. Jun 2018 at 23:42 gailmargolis76 ***@***.***> wrote: > My input is such that each subject has their own graph. This is different > from the example given in train.py where there is only 1 graph (a citation > network). In the tensorflow implementation of gcn, you suggest doing > graph-level classification by combining the adjacency matrices of all the > graphs in the input sample into one large adjacency matrix (as a sparse > block-diagonal matrix). The part I am not sure how to implement in keras is > the pooling of the output to produce 1 classification per graph. Any tips > would be greatly appreciated! > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#33>, or mute the thread > <https://github.com/notifications/unsubscribe-auth/AHAcYJWmmhWbZ80CLdpev3L8aMxNscgTks5uBApogaJpZM4U6flP> > . >