How to collect unsupervised graphSAGE node vectors

[rehmliang]

Hi, I'm training graphSAGE in unsupervised manner to get node representations

In examples/graph_sage_unsup.py ,

    def full_forward(self, x, edge_index):
        for i, conv in enumerate(self.convs):
            x = conv(x, edge_index)
            if i != self.num_layers - 1:
                x = x.relu()
                x = F.dropout(x, p=0.5, training=self.training)
        return x

Does full_forward collect the final node embedding ?

[rusty1s]

Yes, it will return the final node embeddings for all nodes in your graph (without any sampling).