Questions about DGL HeteroGraphs that you never dared to ask

Questions
#41

As suggested here, if you use gcn for aggregator_type, SAGEConv expects the feature shape of source and destination nodes to be the same. You can choose to use a different aggregator type.

#42

Thanks for pointing out the documentation, I did not notice this particularity. It now works, but I am having trouble with custom layers. I am trying to build a HeteroGraphConv with custom layers, but again, my dimensions are not matching.

Here is my customer Test_layer:

class Test_layer(nn.Module): #Replaces SAGEConv layer
    def reset_parameters(self):
        """Reinitialize learnable parameters."""
        gain = nn.init.calculate_gain('relu')
        nn.init.xavier_uniform_(self.fc_self.weight, gain=gain)
        nn.init.xavier_uniform_(self.fc_neigh.weight, gain=gain)
    
    
    def __init__(self,
             in_feats,
             out_feats):
        super(Test_layer, self).__init__()

        self._in_src_feats, self._in_dst_feats = dgl.utils.expand_as_pair(in_feats)
        self._out_feats = out_feats

        # NN through which the initial 'h' state passes
        self.fc_self = nn.Linear(self._in_dst_feats, out_feats)

        # NN through which the neighborhood messages passes
        self.fc_neigh = nn.Linear(self._in_src_feats, out_feats)
        self.reset_parameters()
            
    
    def forward(self, graph, feat):
               
        h_self = feat
        
        #get input features
        graph.srcdata['h'] = h_self
        
        #update nodes values
        graph.update_all( 
            fn.copy_src('h', 'm'), 
            fn.mean('m', 'neigh')) 
        
        h_neigh = graph.dstdata['neigh'] 
    
        print(h_self.shape)
        print(h_neigh.shape)

        # result is output of NN for h_self + output of NN for h_neigh
        rst = self.fc_self(h_self) + self.fc_neigh(h_neigh)
        
        self.reset_parameters()
        
        return rst

Then, I call the HeteroGraph with some user and item features.

layer = dglnn.HeteroGraphConv({'buys':Test_layer(5,100),
                               'bought-by':Test_layer(3,100)},
                               aggregate='sum')

item_feat = torch.randn((g.number_of_nodes('item'), 3))
user_feat = torch.randn((g.number_of_nodes('user'), 5))
features = {'item' : item_feat, 'user' : user_feat }

out = layer(g, features)

I get the following error.
RuntimeError: The size of tensor a (1892) must match the size of tensor b (2492) at non-singleton dimension 0
Which makes sense, since my h_self is of shape torch.Size([1892, 5]) and my h_neigh of size torch.Size([2492, 5]).

How can I arrange this so that my h_neigh becomes the right shape, i.e. 1892 as well? I though that my reduce function fn_mean would do so, but it appears I am not making it work properly.

Again, thanks for your answers, newbie here trying to understand better heterographs in DGL!

#43

Hi @jedbl ,

Could you please open a new post for your question? It’s a bit hard to track under this long thread. Thanks!

#44

Hi @VoVAllen,
Thanks for pointing this out. I created a new trend here.

Thanks!

#45

Hi,

This thread is closed now. If you have any further questions, please feel free to open a new post.

#46