Question about graph sage and role of message_and_aggregate function

[gebrahimi91]

For GraphSage, there is this part in the source code:

 def message_and_aggregate(self, adj_t: SparseTensor,
                          x: OptPairTensor) -> Tensor:

    adj_t = adj_t.set_value(None, layout=None)
    return matmul(adj_t, x[0], reduce=self.aggr)

Can i ask what is adj_t here? and why x is OptPairTensor? I also do not know when and how this function is executed. I tried to print adj_t, but it does not do anything for me. Even printing a random sentence does not happen. May I know how the propagate function executes the message and message_and_aggregate function? Thanks

[rusty1s]

Should be hopefully already answered by #2412 :)