Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP][Feature] DGL Pooling modules #669

wants to merge 21 commits into
base: master


None yet
2 participants
Copy link

commented Jun 18, 2019


Global Pooling(readout)

  • Sum pooling
  • Avg pooling
  • Max pooling
  • Global Attention Pooling
  • Set2Set
  • SortPooling
  • Set Transformer

Sequential Pooling

  • Diffpool

Other issues

  • Pooling on edges
  • MXNet support
  • Performance checking
  • Graph U-Net


Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [$CATEGORY] (such as [Model], [Doc], [Feature]])
  • Changes are complete (i.e. I finished coding on this PR)
  • All changes have test coverage
  • Code is well-documented
  • To the my best knowledge, examples are either not affected by this change,
    or have been fixed to be compatible with this change
  • Related issue is referred in this PR


  • dgl.nn.pytorch.glob

@yzh119 yzh119 referenced this pull request Jun 24, 2019


[Roadmap] DGL support checklist #5

0 of 3 tasks complete

@yzh119 yzh119 marked this pull request as ready for review Jul 4, 2019

yzh119 added some commits Jul 5, 2019

@@ -112,12 +115,36 @@ def copy_to(input, ctx):
def sum(input, dim):
return nd.sum(input, axis=dim)

def all_sum(input):
return input.max()

This comment has been minimized.

Copy link

BarclayII Jul 17, 2019



Top-k features of the given graph with shape :math:`(K, D)`,
if the input graph is a BatchedDGLGraph a tensor with shape
:math:`(B, K, D)` we be returned, where math`B` is the batch

This comment has been minimized.

Copy link

BarclayII Jul 17, 2019


Should we also return the index of the top-k features?


This comment has been minimized.

Copy link

commented Jul 17, 2019

Also I would suggest the backend function addition be separated into a new PR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.