N-hop convlayers & multiple layers in one individual convlayer of the N-hop.

[Jeriousman]

As GNN is propagating messages by going forward with hops, it is different from going deep in general DNN layers.

So my questions are as below:

1. The example code below has 2 SAGEConv layers. Does that mean messages are propagated by 2 hops with these 2 convlayers?

class GNNEncoder(torch.nn.Module):
    def __init__(self, hidden_channels, out_channels):
        super().__init__()
        self.conv1 = SAGEConv((-1, -1), hidden_channels)
        self.conv2 = SAGEConv((-1, -1), out_channels)

    def forward(self, x, edge_index):
        x = self.conv1(x, edge_index).relu()
        x = self.conv2(x, edge_index)
        return x

2. Then if I want to make one individual hop conv deeper (multiple stacked layers like in general dnn), what should I do? Should modify the SAGEConv class? specifically forward part of SAGEConv? (see a part of SAGEConv code below)

Original SAGEConv

 out = self.propagate(edge_index, x=x, size=size)
        out = self.lin_l(out)

Modified SAGEConv for deeper stacked (here 5) layers

 out = self.propagate(edge_index, x=x, size=size)
        out = self.lin_1(out)
        out = self.lin_2(out)
        out = self.lin_3(out)
        out = self.lin_4(out)
        out = self.lin_5(out)

Thank you for you advice in advance.