SAGEConv doc maybe inaccurate?

In the doc of forward func for SAGEConv:

Returns: The output feature of shape (𝑁,π·π‘œπ‘’π‘‘) where π·π‘œπ‘’π‘‘ is size of output feature.

N is the number of nodes in the graph.
However, this may be wrong for block sampled by dataloader:

import dgl
import torch as th
from dgl.nn import SAGEConv

g = dgl.graph(([0, 2, 0, 1, 3, 4], [1, 1, 2, 0, 1, 0]))
train_nid = th.tensor([1, 0, 2, 3], dtype=th.int64)
sampler = dgl.dataloading.MultiLayerNeighborSampler([3])

dataloader = dgl.dataloading.NodeDataLoader(
    g, train_nid, sampler,
    batch_size=2, shuffle=True, drop_last=False, num_workers=1)

conv = SAGEConv(10, 2, 'pool')
feat = th.nn.Embedding(5, 10)

for input_nodes, output_nodes, blocks in dataloader:
    print(conv(blocks[0], feat(input_nodes)))

The returned shape should be (𝑁_dst,π·π‘œπ‘’π‘‘).

Hi, thanks for pointing this out. It is indeed an inaccurate description. Would you mind pushing a fix for us? Thanks!

I find the related doc in python\dgl\nn\pytorch\conv\ So am I just need to edit the doc in this file and submit a pull request?

That’s correct. The doc will be built automatically.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.